FL_Embedding_Classifier.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HvqlPD_IG-ea",
    "outputId": "d09a3a5e-43d1-42bb-de1a-43ad6ed0bb88"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: sentence_transformers in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (3.2.0)\n",
      "Requirement already satisfied: transformers<5.0.0,>=4.41.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sentence_transformers) (4.45.2)\n",
      "Requirement already satisfied: tqdm in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sentence_transformers) (4.66.5)\n",
      "Requirement already satisfied: torch>=1.11.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sentence_transformers) (2.4.1)\n",
      "Requirement already satisfied: scikit-learn in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sentence_transformers) (1.5.2)\n",
      "Requirement already satisfied: scipy in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sentence_transformers) (1.14.1)\n",
      "Requirement already satisfied: huggingface-hub>=0.20.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sentence_transformers) (0.25.2)\n",
      "Requirement already satisfied: Pillow in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sentence_transformers) (10.4.0)\n",
      "Requirement already satisfied: filelock in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from huggingface-hub>=0.20.0->sentence_transformers) (3.16.1)\n",
      "Requirement already satisfied: fsspec>=2023.5.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from huggingface-hub>=0.20.0->sentence_transformers) (2024.9.0)\n",
      "Requirement already satisfied: packaging>=20.9 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from huggingface-hub>=0.20.0->sentence_transformers) (24.1)\n",
      "Requirement already satisfied: pyyaml>=5.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from huggingface-hub>=0.20.0->sentence_transformers) (6.0.2)\n",
      "Requirement already satisfied: requests in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from huggingface-hub>=0.20.0->sentence_transformers) (2.32.3)\n",
      "Requirement already satisfied: typing-extensions>=3.7.4.3 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from huggingface-hub>=0.20.0->sentence_transformers) (4.12.2)\n",
      "Requirement already satisfied: sympy in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from torch>=1.11.0->sentence_transformers) (1.13.3)\n",
      "Requirement already satisfied: networkx in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from torch>=1.11.0->sentence_transformers) (3.4.1)\n",
      "Requirement already satisfied: jinja2 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from torch>=1.11.0->sentence_transformers) (3.1.4)\n",
      "Requirement already satisfied: colorama in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tqdm->sentence_transformers) (0.4.6)\n",
      "Requirement already satisfied: numpy>=1.17 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from transformers<5.0.0,>=4.41.0->sentence_transformers) (1.26.4)\n",
      "Requirement already satisfied: regex!=2019.12.17 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from transformers<5.0.0,>=4.41.0->sentence_transformers) (2024.9.11)\n",
      "Requirement already satisfied: safetensors>=0.4.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from transformers<5.0.0,>=4.41.0->sentence_transformers) (0.4.5)\n",
      "Requirement already satisfied: tokenizers<0.21,>=0.20 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from transformers<5.0.0,>=4.41.0->sentence_transformers) (0.20.1)\n",
      "Requirement already satisfied: joblib>=1.2.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from scikit-learn->sentence_transformers) (1.4.2)\n",
      "Requirement already satisfied: threadpoolctl>=3.1.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from scikit-learn->sentence_transformers) (3.5.0)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from jinja2->torch>=1.11.0->sentence_transformers) (3.0.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests->huggingface-hub>=0.20.0->sentence_transformers) (3.4.0)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests->huggingface-hub>=0.20.0->sentence_transformers) (3.10)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests->huggingface-hub>=0.20.0->sentence_transformers) (2.2.3)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests->huggingface-hub>=0.20.0->sentence_transformers) (2024.8.30)\n",
      "Requirement already satisfied: mpmath<1.4,>=1.1.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from sympy->torch>=1.11.0->sentence_transformers) (1.3.0)\n",
      "Requirement already satisfied: pandas in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (2.2.3)\n",
      "Requirement already satisfied: numpy>=1.23.2 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from pandas) (1.26.4)\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from pandas) (2.9.0.post0)\n",
      "Requirement already satisfied: pytz>=2020.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from pandas) (2024.2)\n",
      "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from pandas) (2024.2)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n",
      "Requirement already satisfied: numpy in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (1.26.4)\n",
      "Requirement already satisfied: scikit-learn in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (1.5.2)\n",
      "Requirement already satisfied: numpy>=1.19.5 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from scikit-learn) (1.26.4)\n",
      "Requirement already satisfied: scipy>=1.6.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from scikit-learn) (1.14.1)\n",
      "Requirement already satisfied: joblib>=1.2.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from scikit-learn) (1.4.2)\n",
      "Requirement already satisfied: threadpoolctl>=3.1.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from scikit-learn) (3.5.0)\n",
      "Requirement already satisfied: matplotlib in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (3.9.2)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (1.3.0)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (0.12.1)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (4.54.1)\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (1.4.7)\n",
      "Requirement already satisfied: numpy>=1.23 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (1.26.4)\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (24.1)\n",
      "Requirement already satisfied: pillow>=8 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (10.4.0)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (3.2.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib) (2.9.0.post0)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\n",
      "Requirement already satisfied: seaborn in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (0.13.2)\n",
      "Requirement already satisfied: numpy!=1.24.0,>=1.20 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from seaborn) (1.26.4)\n",
      "Requirement already satisfied: pandas>=1.2 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from seaborn) (2.2.3)\n",
      "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from seaborn) (3.9.2)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.0)\n",
      "Requirement already satisfied: cycler>=0.10 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.12.1)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.54.1)\n",
      "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.7)\n",
      "Requirement already satisfied: packaging>=20.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (24.1)\n",
      "Requirement already satisfied: pillow>=8 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (10.4.0)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.2.0)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.9.0.post0)\n",
      "Requirement already satisfied: pytz>=2020.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from pandas>=1.2->seaborn) (2024.2)\n",
      "Requirement already satisfied: tzdata>=2022.7 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from pandas>=1.2->seaborn) (2024.2)\n",
      "Requirement already satisfied: six>=1.5 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.16.0)\n",
      "Requirement already satisfied: tf-keras in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (2.17.0)\n",
      "Requirement already satisfied: tensorflow<2.18,>=2.17 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tf-keras) (2.17.1)\n",
      "Requirement already satisfied: tensorflow-intel==2.17.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow<2.18,>=2.17->tf-keras) (2.17.1)\n",
      "Requirement already satisfied: absl-py>=1.0.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (2.1.0)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (1.6.3)\n",
      "Requirement already satisfied: flatbuffers>=24.3.25 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (24.3.25)\n",
      "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.6.0)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.2.0)\n",
      "Requirement already satisfied: h5py>=3.10.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.12.1)\n",
      "Requirement already satisfied: libclang>=13.0.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (18.1.1)\n",
      "Requirement already satisfied: ml-dtypes<0.5.0,>=0.3.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.4.1)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.4.0)\n",
      "Requirement already satisfied: packaging in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (24.1)\n",
      "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (4.25.5)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (2.32.3)\n",
      "Requirement already satisfied: setuptools in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (65.5.0)\n",
      "Requirement already satisfied: six>=1.12.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (1.16.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (2.5.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (4.12.2)\n",
      "Requirement already satisfied: wrapt>=1.11.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (1.16.0)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (1.66.2)\n",
      "Requirement already satisfied: tensorboard<2.18,>=2.17 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (2.17.1)\n",
      "Requirement already satisfied: keras>=3.2.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.6.0)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.31.0)\n",
      "Requirement already satisfied: numpy<2.0.0,>=1.23.5 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (1.26.4)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from astunparse>=1.6.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.44.0)\n",
      "Requirement already satisfied: rich in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from keras>=3.2.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (13.9.2)\n",
      "Requirement already satisfied: namex in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from keras>=3.2.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.0.8)\n",
      "Requirement already satisfied: optree in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from keras>=3.2.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.13.0)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests<3,>=2.21.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.4.0)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests<3,>=2.21.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.10)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests<3,>=2.21.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (2.2.3)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from requests<3,>=2.21.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (2024.8.30)\n",
      "Requirement already satisfied: markdown>=2.6.8 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorboard<2.18,>=2.17->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.7)\n",
      "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorboard<2.18,>=2.17->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.7.2)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from tensorboard<2.18,>=2.17->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.0.4)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from werkzeug>=1.0.1->tensorboard<2.18,>=2.17->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.0.1)\n",
      "Requirement already satisfied: markdown-it-py>=2.2.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from rich->keras>=3.2.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (3.0.0)\n",
      "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from rich->keras>=3.2.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (2.18.0)\n",
      "Requirement already satisfied: mdurl~=0.1 in c:\\users\\nikhi\\desktop\\github projects\\federated-medembedx\\.venv\\lib\\site-packages (from markdown-it-py>=2.2.0->rich->keras>=3.2.0->tensorflow-intel==2.17.1->tensorflow<2.18,>=2.17->tf-keras) (0.1.2)\n"
     ]
    }
   ],
   "source": [
    "# Install the libraries\n",
    "!pip install sentence_transformers\n",
    "!pip install pandas\n",
    "!pip install numpy\n",
    "!pip install scikit-learn\n",
    "!pip install matplotlib\n",
    "!pip install seaborn\n",
    "!pip install tf-keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1fxIGnXgHY1_",
    "outputId": "2eb2b7cb-f661-43f4-f525-6f2ee7f54d31"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\sentence_transformers\\cross_encoder\\CrossEncoder.py:13: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from tqdm.autonotebook import tqdm, trange\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From c:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tf_keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Import the packages\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import os\n",
    "import json\n",
    "from sklearn.manifold import TSNE\n",
    "import numpy as np\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, roc_auc_score, confusion_matrix, recall_score\n",
    "from sklearn.model_selection import cross_validate\n",
    "from sentence_transformers import SentenceTransformer\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.losses import BinaryCrossentropy\n",
    "from tensorflow.keras.metrics import AUC\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "78g-W3kJsAx-"
   },
   "source": [
    "# Federated Learning with Surgical Aggregation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 528,
     "referenced_widgets": [
      "a30e36f3bf9e4c3e85df561d3f1a81a1",
      "63dc8025965242e09dfb7fffb70c953a",
      "4b8aa7e353f045dd8297de8c688b9d40",
      "0cdbc7260ad346048a87a3f8ba22486e",
      "59f57f5ddfd24b53a4f1621f1a6389bb",
      "280332c72db54753a188f6abf68eaaa9",
      "f7d8f5d7761541ec95481008339f2420",
      "2d46775c61384019ad3479eab413e7a9",
      "25d29337a9544ee98ba83c7e9d75f853",
      "bf354c1bbf524f97a07cd2a6c854e22c",
      "199a73af56a5402e9ad15644a7c72755",
      "874a74254ebc4d71a119f67e137c0551",
      "2fbd5496ad3944fd92a3c28f8d3e31a1",
      "a236061375e947dc9612c24adbe4ff0b",
      "cd89e860240142e28da826f6f49ba240",
      "9c8aa38595dc4b7e8ca640d03cc19694",
      "76496dac60c44c839bb441f2f4ff6f9f",
      "57453cbecc0b4d7ea786935449271c90",
      "6bdcd35840594df4ab40d4c64e660dba",
      "f3384eb9244c4701ac8c1b16772132cf",
      "f412568cd7f741c5a2d4a18416981739",
      "50b11ecfa6ee4d4d8d4fc97a31a4551b",
      "887871630cd944d89907336e3124835e",
      "412ec4ec63d448939556ff9f40e5805b",
      "2d94a369fc234cbc8452df68917e75b6",
      "04ff5f37db7c42f2980f807bc3032586",
      "d357f7f4bd3f4af7af517dd7a725d2f0",
      "a53e2fc70d75494da4d6106f6608875b",
      "77d4965ca1c1401381aca48ec309280e",
      "5e91ad19f8ad463f9cb5a0f9d1e8c294",
      "1cc1e2d9840149c497d7d90b5dbb2d55",
      "74a45523ed954b0abd9e7a05045d2daf",
      "c2058ed868ff4815a9ef7ea89e85f022",
      "6274cc6165234aaba2c930d23350b362",
      "5e61f9e947714e64a29e116c44a124f0",
      "e2c7cbcbf7cd4e0ca1352925b8c7bdea",
      "6b70db174de04c9ba25282525de05387",
      "41be3437b15c4e57a75534d238a829a5",
      "e48c72f5aa414987a2ce9409e9b5ad69",
      "2d273cb8c27347b68a191ee17dc89311",
      "30a058a01b434739adf517283900f1c8",
      "9036a15be46e45799e8e3cc854fa88bb",
      "8cd133cc3bd44a9cb49c7d1b6a4db642",
      "16d38234182b49dda5d3115621aecb39",
      "590a4e7454af492a8910e8fc1940b872",
      "b73272664dfb4605b1d3a85928865304",
      "52067038eb414f5f983281696bcf84b1",
      "3960d078abed4760b043dd20fddcf949",
      "14822fe23d064263bafda0ecaa573e15",
      "879a060bd57146398f7c175704456532",
      "64a3c90a4ba94283bbc717031722190c",
      "658c15a0430b4a788e9874b236cfb86d",
      "701a5fc9c6ac4efebee807273eae66e1",
      "2803d2f51c944e789952f5e4368c6c06",
      "267f0cdfde1f42cfb5a09d0da0c453ea",
      "932c4d4ac22d402fa5f1564437dc5e5b",
      "35750f3fe46e40b99058d14ca7ed800b",
      "f4e20a4a864b4c13adc38f22fccb80a0",
      "3987a825b2584353a2be546f63911a0c",
      "1bc226c51f954bbf98c13d2310940135",
      "fffef1e3409c4e929c21060b07853091",
      "692fbf8b0b7542a39c3978569db3a9f8",
      "b05eb88e352947e295a0ed84710ed5f0",
      "dcbf618d0a85491f8254ead7a0c027f5",
      "4f3486a446af44019511654af1f51312",
      "14cb9e333a89414b9167d21d26760cff",
      "14261702d08a4d019e8ce8ee1f08d4ce",
      "12e62aa39c8a475ebd65aaf5e5e84338",
      "22223361bfd3496ca312f53e3653cfed",
      "224ab887d94a455cb3664d0c7d7ddde2",
      "81ae5aa2028c46248b01d916cce292fa",
      "89c8e756392047b48e155bc153b4f6dd",
      "9ce5c894beb448b1b5935907c38d9138",
      "fbb3251aed0743e1be86088735d2ba33",
      "22617e8f1b384d17b2481d77edb94658",
      "a02025fa486a4cab95eb79496bc3e7ea",
      "89e43961094244889ca5e14889d9f815",
      "86b9a304e58446939dbb1d7a023de81f",
      "87034250c3894eee883cb3ecb9347078",
      "efa59884be0e4fa6bb6ddf4d4885ab52",
      "3f2a6b57c81c4dc99404363b0b0e5632",
      "60bf8b23189b45949746bed8dbb91765",
      "ab01960dd6954af78441736c8669b241",
      "e3f45f680ec84c4eb6ee402b187e38a8",
      "bd7036844925403fb3db2f1985fbf7f8",
      "33c5f2daa3a146e4ad66a753a4969e71",
      "1e89725dd21a4a28a127279a274819d8",
      "0a66bb0689824b83a0230934c12374e3",
      "71f5342382cb4fdf818c03978a4ccccb",
      "fb9ef2497d3b47a297501004b0176cde",
      "b63ec0fd33c448e89cc03e6000abee00",
      "7da0f13278624d2a91348680b4f9e6be",
      "8a82c40aedbc4555a03a8c29981740dd",
      "2377c5a7810c4e86849f2f1889cb0a5e",
      "4a8445cda576479ab8663568b2e58fe0",
      "2153807b370049bd95c515e53f487f72",
      "291e805751f84bb3a29d25c8b7cb3204",
      "10c6a410b1ca4210812a84b8d1bd9984",
      "c2d096efb1db4dde86fd42ff3a4df605",
      "0ea78511c0d54dd1acbd29baaf0cbf58",
      "abdbb22939fd41db9b0a3ee58c7ed13b",
      "c92270c929c24bcb847c6739686a799b",
      "7b87dd0653c14863a8a2af560d5e7c42",
      "fec6c46a94314fff8f5fa6db76c722a2",
      "7a839c9bc6ca43fdbf6d63cfe823af63",
      "59d57ba6032a46d79628065ec01cff0f",
      "e5aa932cda80483d8328d9f942462e9e",
      "e0e136927c3644c6bfb9d611817870c8",
      "b5a63beb6d2b4249871961e1a0a9b937",
      "f94f4f97ce194f77bdd6d4bdfea4edea",
      "332a06b57849496c9a9b73b0cf2b4737",
      "2c7d422cafe04b1cb2112f28fd631194",
      "56762b63b5c74f56bdeb19012c7e509b",
      "fe74971a69824ace86c03a408afe7100",
      "af86bd6b8bef45c99bc6d61af2c4f71f",
      "6b07c11c0b634256be6c9dc37369fbb5",
      "bf34bae971834a2ca8ab1ac10e833986",
      "b4717e6adc3e42648ffbf1e202a852ca",
      "71cef30c369b4878823b8a4db5b44f5b",
      "a0fd105ea5f046c18b8191d1ae5d89bc",
      "2319ac540cdb44eba7715740e1f391b3"
     ]
    },
    "id": "9KitraTVHf1d",
    "outputId": "448a6dff-f056-4f2b-9b40-b1ea1e3bf8b3"
   },
   "outputs": [],
   "source": [
    "# Load the embedding model\n",
    "model = SentenceTransformer(\"all-MiniLM-L6-v2\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "qaurbTtCGu0q"
   },
   "outputs": [],
   "source": [
    "# Load the datasets and create embeddings\n",
    "file_path = r\"data\\IU-GroundTruth.csv\"\n",
    "report_column = 'REPORT'\n",
    "df = pd.read_csv(file_path)\n",
    "allReports = df[report_column]\n",
    "allEmbeddings = [];\n",
    "\n",
    "for report in allReports:\n",
    "    response = model.encode(report)\n",
    "    allEmbeddings.append(response)\n",
    "\n",
    "label_columns = ['Enlarged Cardiom.','Cardiomegaly','Lung Lesion','Lung Opacity','Edema','Consolidation','Pneumonia','Atelectasis','Pneumothorax','Pleural Effusion','Pleural Other','Fracture','Support Devices']\n",
    "labels = df[label_columns].values\n",
    "\n",
    "allEmbeddings = np.array(allEmbeddings)\n",
    "labels = np.array(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "train_test_splits = [] \n",
    "test_size = 0.2\n",
    "for i in range(20):\n",
    "    rng=random.randint(1, 100)\n",
    "    X_train, X_test, y_train, y_test = train_test_split(allEmbeddings, labels, test_size=test_size, random_state=rng)\n",
    "    train_test_splits.append((X_train, X_test,y_train,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "qNfkcRaZYDEf"
   },
   "outputs": [],
   "source": [
    "# Function to create an MLP classifier\n",
    "def create_mlp_model(input_dim, output_dim):\n",
    "    model = Sequential()\n",
    "\n",
    "    inputs = keras.Input(shape=(input_dim,))\n",
    "    x = layers.Dense(512, activation=\"relu\")(inputs)\n",
    "    x = layers.Dropout(0.2)(x) # optional dropout\n",
    "    x = layers.Dense(256, activation=\"relu\")(x)\n",
    "    x = layers.Dropout(0.2)(x) # optional dropout\n",
    "    outputs = layers.Dense(output_dim, activation=\"sigmoid\")(x)\n",
    "    model = keras.Model(inputs=inputs, outputs=outputs)\n",
    "\n",
    "\n",
    "\n",
    "    # Compile model\n",
    "    model.compile(optimizer=Adam(),\n",
    "                  loss=BinaryCrossentropy(),\n",
    "                  metrics=[AUC(name='auc')])  # Use AUC as a metric\n",
    "\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "7orWZVCppQR-"
   },
   "outputs": [],
   "source": [
    "# Surgical Aggregation\n",
    "\n",
    "def surgical_aggregation(nodes, max_label, global_model):\n",
    "\n",
    "  \"\"\"\n",
    "  This function will surgically aggregate the last layer\n",
    "  with the assumption that all the other layers\n",
    "  are part of the representation block.\n",
    "\n",
    "  Args:\n",
    "    nodes: list of nodes with their models and label indices\n",
    "    max_label: maximum number of labels / maximum label indice\n",
    "    global_model: global model to update\n",
    "\n",
    "  \"\"\"\n",
    "\n",
    "  # Extract the models and label indices from each node\n",
    "  models = [node['classifier'] for node in nodes]\n",
    "  labels = [node['label_names'] for node in nodes]\n",
    "\n",
    "  # Identify the common and unique classes across all the models\n",
    "\n",
    "  # Create a global final layer with all the labels\n",
    "  lw = models[0].layers[-1].get_weights() # Get the last layer from one of the models\n",
    "  row_shape = lw[0].shape[0] # Get the shape of the last layer\n",
    "  final_layer_w = np.zeros((lw[0].shape[0],max_label+1)) # Use the last layer's shape to initialize the final layer for the global model\n",
    "  final_layer_b = np.zeros(max_label+1) # Same for the bias\n",
    "\n",
    "  node_count = np.zeros(max_label+1) # Number of contributing to each label\n",
    "\n",
    "\n",
    "\n",
    "  num_layers = len(models[0].layers) # Get the number of layers\n",
    "  for iter,layer in enumerate(models[0].layers): # iteratre through all the layers\n",
    "\n",
    "    if isinstance(layer, tf.keras.layers.Dense): # Check if the layer type is a dense layer\n",
    "\n",
    "      if iter == num_layers-1: # Check if this is the final layer (for surgical aggregation)\n",
    "\n",
    "        # Surgically Aggregate the final layer\n",
    "\n",
    "\n",
    "        node_count = np.zeros(max_label+1) # Initialize an array to count the number of nodes contributing to each label\n",
    "        for iter1,model in enumerate(models): # iterate through all the models\n",
    "          w = (model.layers[iter].get_weights()) # get the weights for the last layer\n",
    "          w_0 = w[0] # get the weight matrix\n",
    "          w_1 = w[1] # get the bias array\n",
    "\n",
    "          for l_iter,l in enumerate(labels[iter1]): #iterate through all the labels in this model\n",
    "\n",
    "            final_layer_w[:,l] += w_0[:,l_iter] # FedAvg on each label this model is contributing to\n",
    "            final_layer_b[l] += w_1[l_iter]\n",
    "            node_count[l] += 1\n",
    "\n",
    "        for l in range(max_label+1):\n",
    "          if node_count[l] > 0: # if this label has atleast one contributor\n",
    "            final_layer_w[:,l] /= node_count[l] # perform fedavg\n",
    "            final_layer_b[l] /= node_count[l] # perform fedavg\n",
    "\n",
    "        for iter1,model in enumerate(models):\n",
    "          fw = final_layer_w[:,labels[iter1]] # reassign the weights to each model\n",
    "          fb = final_layer_b[labels[iter1]]\n",
    "          model.layers[iter].set_weights([fw,fb])\n",
    "        global_model.layers[iter].set_weights([final_layer_w,final_layer_b]) # assign the weights to the global model\n",
    "\n",
    "      else:\n",
    "        # Normal FedAvg for all other layers\n",
    "        d_w = [] # for weights\n",
    "        d_b = [] # for bias\n",
    "        for model in models:\n",
    "          d1 = model.layers[iter].get_weights()\n",
    "          d_w.append(d1[0])\n",
    "          d_b.append(d1[1])\n",
    "\n",
    "        new_w = np.mean(d_w,axis=0)\n",
    "        new_b = np.mean(d_b,axis=0)\n",
    "        new_d = [new_w,new_b]\n",
    "\n",
    "        for model in models:\n",
    "          model.layers[iter].set_weights(new_d)\n",
    "        global_model.layers[iter].set_weights(new_d)\n",
    "\n",
    "  # Reassign the models to nodes\n",
    "  for iter,node in enumerate(nodes):\n",
    "    node['classifier'] = models[iter]\n",
    "\n",
    "  return global_model, nodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "25u_CTeKIoXo"
   },
   "outputs": [],
   "source": [
    "# Function to create the datasets for FL\n",
    "def create_FL_Datasets(X_train, y_train, n_nodes, labels_shared):\n",
    "\n",
    "\n",
    "  # Initialize the sample and label indices\n",
    "  sample_indices = np.arange(0, X_train.shape[0])\n",
    "  label_indices = np.arange(0, y_train.shape[1])\n",
    "\n",
    "  # Let's shuffle the indices\n",
    "  np.random.shuffle(label_indices)\n",
    "  np.random.shuffle(sample_indices)\n",
    "\n",
    "  # Picking the shared labels\n",
    "  if labels_shared > 0:\n",
    "    shared_labels = label_indices[:labels_shared]\n",
    "    label_indices = label_indices[labels_shared:]\n",
    "  else:\n",
    "    shared_labels = []\n",
    "\n",
    "\n",
    "\n",
    "  # Split the remaining labels into different nodes.\n",
    "  label_indices_per_node = np.array_split(label_indices, n_nodes)\n",
    "\n",
    "\n",
    "  # Split the dataset into different nodes\n",
    "  sample_indices_per_node = np.array_split(sample_indices, n_nodes)\n",
    "\n",
    "  # Let's create the nodes. Each node will store a dictionary of X_train, y_train,\n",
    "  # and label_names\n",
    "  nodes = []\n",
    "  for i in range(n_nodes):\n",
    "    node = {}\n",
    "    node['X_train'] = X_train[sample_indices_per_node[i]]\n",
    "    if len(shared_labels) > 0:\n",
    "      node['y_train'] = y_train[np.ix_(sample_indices_per_node[i],np.concatenate((shared_labels,label_indices_per_node[i])))]\n",
    "      node['label_names'] = np.concatenate((shared_labels,label_indices_per_node[i]))\n",
    "    else:\n",
    "      node['y_train'] = y_train[np.ix_(sample_indices_per_node[i],label_indices_per_node[i])]\n",
    "      node['label_names'] = label_indices_per_node[i]\n",
    "\n",
    "    node['classifier'] = create_mlp_model(node['X_train'].shape[1], node['y_train'].shape[1])\n",
    "    nodes.append(node)\n",
    "\n",
    "  return nodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "rWqjLQkUMepG"
   },
   "outputs": [],
   "source": [
    "def get_metrics(model, X_test, y_test, labels):\n",
    "    import numpy as np\n",
    "    from sklearn.metrics import accuracy_score, recall_score, roc_auc_score, roc_curve\n",
    "\n",
    "    # Predict probabilities\n",
    "    y_pred_proba = model.predict(X_test)\n",
    "    y_true = y_test.copy()\n",
    "\n",
    "    # Metrics calculation\n",
    "    metrics = {}\n",
    "    for i in range(labels.shape[1]):\n",
    "        # Calculate the ROC curve to find optimal threshold\n",
    "        fpr, tpr, thresholds = roc_curve(y_true[:, i], y_pred_proba[:, i])\n",
    "        # Calculate Youden's J statistic for each threshold\n",
    "        J = tpr - fpr\n",
    "        optimal_idx = np.argmax(J)\n",
    "        optimal_threshold = thresholds[optimal_idx]\n",
    "\n",
    "        # Apply the optimal threshold to get binary predictions\n",
    "        y_pred = (y_pred_proba[:, i] >= optimal_threshold).astype(int)\n",
    "\n",
    "        # Accuracy\n",
    "        accuracy = accuracy_score(y_true[:, i], y_pred)\n",
    "\n",
    "        # Sensitivity (Recall for the positive class)\n",
    "        sensitivity = recall_score(y_true[:, i], y_pred, pos_label=1)\n",
    "\n",
    "        # Specificity: Recall for the negative class (pos_label=0)\n",
    "        specificity = recall_score(y_true[:, i], y_pred, pos_label=0)\n",
    "\n",
    "        # AUC (Area Under the ROC Curve)\n",
    "        auc = roc_auc_score(y_true[:, i], y_pred_proba[:, i])\n",
    "\n",
    "        # Store metrics for this label\n",
    "        metrics[f'Label_{i}'] = {\n",
    "            'Optimal Threshold': optimal_threshold,\n",
    "            'Accuracy': accuracy,\n",
    "            'Sensitivity': sensitivity,\n",
    "            'Specificity': specificity,\n",
    "            'AUC': auc\n",
    "        }\n",
    "\n",
    "    # Overall metrics\n",
    "    metrics['Overall'] = {\n",
    "        'Accuracy': np.mean([m['Accuracy'] for m in metrics.values() if isinstance(m, dict)]),\n",
    "        'Sensitivity': np.mean([m['Sensitivity'] for m in metrics.values() if isinstance(m, dict)]),\n",
    "        'Specificity': np.mean([m['Specificity'] for m in metrics.values() if isinstance(m, dict)]),\n",
    "        'AUC': np.mean([m['AUC'] for m in metrics.values() if isinstance(m, dict)])\n",
    "    }\n",
    "\n",
    "    return metrics\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def old_get_metrics(model, X_test, y_test, labels, threshold=0.5):\n",
    "        # Predict probabilities\n",
    "        y_pred_proba = model.predict(X_test)\n",
    "        \n",
    "        # Apply consistent threshold\n",
    "        y_pred = (y_pred_proba >= threshold).astype(int)\n",
    "        y_true = y_test.copy()\n",
    "\n",
    "        # Metrics calculation\n",
    "        metrics = {}\n",
    "        for i in range(labels.shape[1]):\n",
    "            metrics[f'Label_{i}'] = {\n",
    "                'Accuracy': accuracy_score(y_true[:, i], y_pred[:, i]),\n",
    "                'Sensitivity': recall_score(y_true[:, i], y_pred[:, i], pos_label=1),\n",
    "                'Specificity': recall_score(y_true[:, i], y_pred[:, i], pos_label=0),\n",
    "                'AUC': roc_auc_score(y_true[:, i], y_pred_proba[:, i])\n",
    "            }\n",
    "        \n",
    "        # Overall metrics\n",
    "        metrics['Overall'] = {\n",
    "            'Accuracy': np.mean([m['Accuracy'] for m in metrics.values() if isinstance(m, dict)]),\n",
    "            'Sensitivity': np.mean([m['Sensitivity'] for m in metrics.values() if isinstance(m, dict)]),\n",
    "            'AUC': np.mean([m['AUC'] for m in metrics.values() if isinstance(m, dict)])\n",
    "        }\n",
    "        \n",
    "        return metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "lz24rDr6T5zt"
   },
   "source": [
    "# Experiment 1: Baseline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_serializable(obj):\n",
    "    if isinstance(obj, dict):\n",
    "        return {k: make_serializable(v) for k, v in obj.items()}\n",
    "    elif isinstance(obj, list):\n",
    "        return [make_serializable(v) for v in obj]\n",
    "    elif isinstance(obj, np.ndarray):\n",
    "        return obj.tolist()\n",
    "    elif isinstance(obj, np.integer):\n",
    "        return int(obj)\n",
    "    elif isinstance(obj, np.floating):\n",
    "        return float(obj)\n",
    "    else:\n",
    "        return obj\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running iteration 1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "Running iteration 3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 14...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 15...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 16...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 17...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 18...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 19...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 20...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "All 20 iterations are complete, and results are saved.\n"
     ]
    }
   ],
   "source": [
    "# Create the output directory if it doesn't exist\n",
    "output_dir = 'data/Experiment 1'\n",
    "if not os.path.exists(output_dir):\n",
    "    os.makedirs(output_dir)\n",
    "\n",
    "# Experiment parameters\n",
    "num_iterations = 20\n",
    "num_rounds = 20\n",
    "num_epochs = 1\n",
    "labels_shared = 0\n",
    "test_size = 0.2\n",
    "\n",
    "all_experiment_results = []\n",
    "\n",
    "for iteration in range(num_iterations):\n",
    "    print(f\"Running iteration {iteration + 1}...\")\n",
    "\n",
    "    # Initialize containers to store results for this iteration\n",
    "    iteration_global_models = []\n",
    "    iteration_perf = []\n",
    "    iteration_label_names = []\n",
    "    iteration_metrics = []\n",
    "\n",
    "    #Loadind up train test split\n",
    "    X_train, X_test, y_train, y_test = train_test_splits[iteration]\n",
    "\n",
    "    base_model = create_mlp_model(allEmbeddings.shape[1], labels.shape[1])\n",
    "    base_model.fit(X_train, y_train, epochs=20, verbose=0)\n",
    "    base_perf = base_model.evaluate(X_test,y_test, verbose=0)\n",
    "    iteration_global_models.append(base_model)\n",
    "    iteration_perf.append(base_perf)\n",
    "\n",
    "    base_metrics = get_metrics(base_model, X_test, y_test, labels)\n",
    "    iteration_metrics.append(base_metrics)\n",
    "\n",
    "    # Convert data to JSON-serializable format\n",
    "    iteration_perf_serializable = make_serializable(iteration_perf)\n",
    "    iteration_label_names_serializable = make_serializable(iteration_label_names)\n",
    "    iteration_metrics_serializable = make_serializable(iteration_metrics)\n",
    "\n",
    "    # Store results of this iteration in a specific folder\n",
    "    iteration_dir = os.path.join(output_dir, f'iteration_{iteration + 1}')\n",
    "    os.makedirs(iteration_dir, exist_ok=True)\n",
    "\n",
    "    # Save performance metrics as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_perf.json'), 'w') as f:\n",
    "        json.dump(iteration_perf_serializable, f)\n",
    "\n",
    "    # Save label names as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_label_names.json'), 'w') as f:\n",
    "        json.dump(iteration_label_names_serializable, f)\n",
    "\n",
    "    # Save detailed metrics as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_metrics.json'), 'w') as f:\n",
    "        json.dump(iteration_metrics_serializable, f)\n",
    "\n",
    "    # Save global models\n",
    "    models_dir = os.path.join(iteration_dir, 'models')\n",
    "    os.makedirs(models_dir, exist_ok=True)\n",
    "    for idx, model in enumerate(iteration_global_models):\n",
    "        # Save model architecture to JSON\n",
    "        model_json = model.to_json()\n",
    "        with open(os.path.join(models_dir, f'global_model_{idx}.json'), 'w') as json_file:\n",
    "            json_file.write(model_json)\n",
    "        # Save model weights (weights cannot be saved as JSON)\n",
    "        model.save_weights(os.path.join(models_dir, f'global_model_{idx}.weights.h5'))\n",
    "    \n",
    "    # Save iteration-level data for comparison\n",
    "    experiment_data = {\n",
    "        'perf': iteration_perf_serializable,\n",
    "        'label_names': iteration_label_names_serializable,\n",
    "        'metrics': iteration_metrics_serializable\n",
    "    }\n",
    "    all_experiment_results.append(experiment_data)\n",
    "\n",
    "# Save all experiment results as JSON\n",
    "all_experiment_results_serializable = make_serializable(all_experiment_results)\n",
    "with open(os.path.join(output_dir, 'all_experiment_results.json'), 'w') as f:\n",
    "    json.dump(all_experiment_results_serializable, f)\n",
    "\n",
    "print(\"All 20 iterations are complete, and results are saved.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the directory where the experiment data is stored\n",
    "experiment_dir = 'data/Experiment 1'\n",
    "\n",
    "# Load the overall results of all experiments\n",
    "with open(os.path.join(experiment_dir, 'all_experiment_results.json'), 'r') as f:\n",
    "    all_experiment_results = json.load(f)\n",
    "\n",
    "\n",
    "all_auc = []      # To store AUC values across iterations\n",
    "all_accuracy = [] # To store accuracy values across iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "for experiment in all_experiment_results:\n",
    "    perf_metrics = experiment['perf']\n",
    "    auc_values = [perf[1] for perf in perf_metrics]  # Assuming AUC is at index 1 in 'perf'\n",
    "    # Extract 'Accuracy' from 'metrics' dictionary for each node\n",
    "    accuracy_values = [metric['Overall']['Accuracy'] for metric in experiment['metrics']]\n",
    "    \n",
    "    all_auc.append(auc_values)\n",
    "    all_accuracy.append(accuracy_values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert lists to NumPy arrays for easy calculation of mean and std dev\n",
    "all_auc = np.array(all_auc)\n",
    "all_accuracy = np.array(all_accuracy)\n",
    "\n",
    "# Calculate mean and standard deviation for AUC and Accuracy across all iterations\n",
    "mean_auc = np.mean(all_auc, axis=0)\n",
    "std_auc = np.std(all_auc, axis=0)\n",
    "\n",
    "mean_accuracy = np.mean(all_accuracy, axis=0)\n",
    "std_accuracy = np.std(all_accuracy, axis=0)\n",
    "\n",
    "# Define the number of nodes used during the experiment\n",
    "n_nodes = 0  # Node range from 2 to 10 inclusive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "'x' and 'y' must have the same size",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[19], line 3\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# Plotting Mean AUC with Standard Deviation\u001b[39;00m\n\u001b[0;32m      2\u001b[0m plt\u001b[38;5;241m.\u001b[39mfigure(figsize\u001b[38;5;241m=\u001b[39m(\u001b[38;5;241m10\u001b[39m, \u001b[38;5;241m5\u001b[39m))\n\u001b[1;32m----> 3\u001b[0m \u001b[43mplt\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43merrorbar\u001b[49m\u001b[43m(\u001b[49m\u001b[43mn_nodes\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmean_auc\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43myerr\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstd_auc\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfmt\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43m-o\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcapsize\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m5\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlabel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mAUC\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[0;32m      4\u001b[0m \u001b[38;5;66;03m# Base model is a constant baseline (no error bars needed):\u001b[39;00m\n\u001b[0;32m      5\u001b[0m plt\u001b[38;5;241m.\u001b[39mplot(\n\u001b[0;32m      6\u001b[0m     n_nodes,\n\u001b[0;32m      7\u001b[0m     [base_perf[\u001b[38;5;241m1\u001b[39m]] \u001b[38;5;241m*\u001b[39m \u001b[38;5;28mlen\u001b[39m(n_nodes),\n\u001b[0;32m      8\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mr--\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m      9\u001b[0m     label\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mBase Model\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m     10\u001b[0m )\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\matplotlib\\pyplot.py:3246\u001b[0m, in \u001b[0;36merrorbar\u001b[1;34m(x, y, yerr, xerr, fmt, ecolor, elinewidth, capsize, barsabove, lolims, uplims, xlolims, xuplims, errorevery, capthick, data, **kwargs)\u001b[0m\n\u001b[0;32m   3225\u001b[0m \u001b[38;5;129m@_copy_docstring_and_deprecators\u001b[39m(Axes\u001b[38;5;241m.\u001b[39merrorbar)\n\u001b[0;32m   3226\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21merrorbar\u001b[39m(\n\u001b[0;32m   3227\u001b[0m     x: \u001b[38;5;28mfloat\u001b[39m \u001b[38;5;241m|\u001b[39m ArrayLike,\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m   3244\u001b[0m     \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs,\n\u001b[0;32m   3245\u001b[0m ) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m ErrorbarContainer:\n\u001b[1;32m-> 3246\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mgca\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43merrorbar\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m   3247\u001b[0m \u001b[43m        \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3248\u001b[0m \u001b[43m        \u001b[49m\u001b[43my\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3249\u001b[0m \u001b[43m        \u001b[49m\u001b[43myerr\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43myerr\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3250\u001b[0m \u001b[43m        \u001b[49m\u001b[43mxerr\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mxerr\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3251\u001b[0m \u001b[43m        \u001b[49m\u001b[43mfmt\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mfmt\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3252\u001b[0m \u001b[43m        \u001b[49m\u001b[43mecolor\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mecolor\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3253\u001b[0m \u001b[43m        \u001b[49m\u001b[43melinewidth\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43melinewidth\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3254\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcapsize\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcapsize\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3255\u001b[0m \u001b[43m        \u001b[49m\u001b[43mbarsabove\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbarsabove\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3256\u001b[0m \u001b[43m        \u001b[49m\u001b[43mlolims\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mlolims\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3257\u001b[0m \u001b[43m        \u001b[49m\u001b[43muplims\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43muplims\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3258\u001b[0m \u001b[43m        \u001b[49m\u001b[43mxlolims\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mxlolims\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3259\u001b[0m \u001b[43m        \u001b[49m\u001b[43mxuplims\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mxuplims\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3260\u001b[0m \u001b[43m        \u001b[49m\u001b[43merrorevery\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43merrorevery\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3261\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcapthick\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcapthick\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3262\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m{\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mdata\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[43m}\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mif\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01mis\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;129;43;01mnot\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43;01melse\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43m{\u001b[49m\u001b[43m}\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3263\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   3264\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\matplotlib\\__init__.py:1473\u001b[0m, in \u001b[0;36m_preprocess_data.<locals>.inner\u001b[1;34m(ax, data, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1470\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(func)\n\u001b[0;32m   1471\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21minner\u001b[39m(ax, \u001b[38;5;241m*\u001b[39margs, data\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[0;32m   1472\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m data \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m-> 1473\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunc\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m   1474\u001b[0m \u001b[43m            \u001b[49m\u001b[43max\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   1475\u001b[0m \u001b[43m            \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;28;43mmap\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43msanitize_sequence\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   1476\u001b[0m \u001b[43m            \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43m{\u001b[49m\u001b[43mk\u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43msanitize_sequence\u001b[49m\u001b[43m(\u001b[49m\u001b[43mv\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mfor\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mk\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mv\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mkwargs\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mitems\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m}\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   1478\u001b[0m     bound \u001b[38;5;241m=\u001b[39m new_sig\u001b[38;5;241m.\u001b[39mbind(ax, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m   1479\u001b[0m     auto_label \u001b[38;5;241m=\u001b[39m (bound\u001b[38;5;241m.\u001b[39marguments\u001b[38;5;241m.\u001b[39mget(label_namer)\n\u001b[0;32m   1480\u001b[0m                   \u001b[38;5;129;01mor\u001b[39;00m bound\u001b[38;5;241m.\u001b[39mkwargs\u001b[38;5;241m.\u001b[39mget(label_namer))\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\matplotlib\\axes\\_axes.py:3633\u001b[0m, in \u001b[0;36mAxes.errorbar\u001b[1;34m(self, x, y, yerr, xerr, fmt, ecolor, elinewidth, capsize, barsabove, lolims, uplims, xlolims, xuplims, errorevery, capthick, **kwargs)\u001b[0m\n\u001b[0;32m   3631\u001b[0m x, y \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39matleast_1d(x, y)  \u001b[38;5;66;03m# Make sure all the args are iterable.\u001b[39;00m\n\u001b[0;32m   3632\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(x) \u001b[38;5;241m!=\u001b[39m \u001b[38;5;28mlen\u001b[39m(y):\n\u001b[1;32m-> 3633\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mx\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m and \u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124my\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m must have the same size\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m   3635\u001b[0m everymask \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_errorevery_to_mask(x, errorevery)\n\u001b[0;32m   3637\u001b[0m label \u001b[38;5;241m=\u001b[39m kwargs\u001b[38;5;241m.\u001b[39mpop(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mlabel\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m)\n",
      "\u001b[1;31mValueError\u001b[0m: 'x' and 'y' must have the same size"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting Mean AUC with Standard Deviation\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.errorbar(n_nodes, mean_auc, yerr=std_auc, fmt='-o', capsize=5, label='AUC')\n",
    "# Base model is a constant baseline (no error bars needed):\n",
    "plt.plot(\n",
    "    n_nodes,\n",
    "    [base_perf[1]] * len(n_nodes),\n",
    "    'r--',\n",
    "    label='Base Model'\n",
    ")\n",
    "\n",
    "plt.xlabel('Number of Nodes')\n",
    "plt.ylabel('Mean AUC ± Std Dev')\n",
    "plt.title('Mean AUC vs Number of Nodes with Standard Deviation')\n",
    "plt.ylim(0.0, 1)\n",
    "plt.xticks(n_nodes)\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 472
    },
    "id": "MvaiwvWb1-EJ",
    "outputId": "b19a946a-570a-4427-8d52-a5aa898857b5"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot AUC vs. number of nodes using seaborn\n",
    "perf_AUC = [x[1] for x in allPerf]\n",
    "n_nodes = range(2,11)\n",
    "plt.figure(figsize=(8, 5))\n",
    "\n",
    "plt.errorbar(\n",
    "    n_nodes,        # x-values\n",
    "    perf_AUC,       # y-values (the means)\n",
    "    yerr=np.std(perf_AUC, axis=0),   # error (standard deviation)\n",
    "    fmt='-o',       # line and marker style\n",
    "    capsize=5,      # adds \"caps\" to the error bars\n",
    "    label='Federated Learning'\n",
    ")\n",
    "\n",
    "# Base model is a constant baseline (no error bars needed):\n",
    "plt.plot(\n",
    "    n_nodes,\n",
    "    [base_perf[1]] * len(n_nodes),\n",
    "    'r--',\n",
    "    label='Base Model'\n",
    ")\n",
    "\n",
    "plt.xlabel('Number of Nodes')\n",
    "plt.ylabel('AUC')\n",
    "plt.title('AUC vs Number of Nodes')\n",
    "plt.ylim(0, 1)\n",
    "plt.xticks(n_nodes)\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 472
    },
    "id": "xkchnL9m3CPw",
    "outputId": "119193fe-4411-49b0-8136-f6fdcf86f77a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Mean Accuracy vs. number of nodes\n",
    "perf_Accuracy = [x['Overall']['Accuracy'] for x in allMetrics]\n",
    "\n",
    "# For the base model\n",
    "base_overall_accuracy = base_metrics['Overall']['Accuracy']\n",
    "\n",
    "plt.figure(figsize=(8, 5))\n",
    "\n",
    "plt.errorbar(\n",
    "    n_nodes,\n",
    "    perf_Accuracy,\n",
    "    yerr=np.std(perf_Accuracy, axis=0),\n",
    "    fmt='-o',\n",
    "    capsize=5,\n",
    "    label='Federated Learning'\n",
    ")\n",
    "\n",
    "plt.plot(\n",
    "    n_nodes,\n",
    "    [base_overall_accuracy] * len(n_nodes),\n",
    "    'r--',\n",
    "    label='Base Model'\n",
    ")\n",
    "\n",
    "plt.xlabel('Number of Nodes')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Accuracy vs Number of Nodes')\n",
    "plt.ylim(0.0, 1.0)\n",
    "plt.xticks(n_nodes)\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Experiment 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Running Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_serializable(obj):\n",
    "    if isinstance(obj, dict):\n",
    "        return {k: make_serializable(v) for k, v in obj.items()}\n",
    "    elif isinstance(obj, list):\n",
    "        return [make_serializable(v) for v in obj]\n",
    "    elif isinstance(obj, np.ndarray):\n",
    "        return obj.tolist()\n",
    "    elif isinstance(obj, np.integer):\n",
    "        return int(obj)\n",
    "    elif isinstance(obj, np.floating):\n",
    "        return float(obj)\n",
    "    else:\n",
    "        return obj\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running iteration 1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 14...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 15...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 16...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 17...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 18...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 19...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "Running iteration 20...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "All 20 iterations are complete, and results are saved.\n"
     ]
    }
   ],
   "source": [
    "# Create the output directory if it doesn't exist\n",
    "output_dir = 'data/Experiment 2'\n",
    "if not os.path.exists(output_dir):\n",
    "    os.makedirs(output_dir)\n",
    "\n",
    "\n",
    "# Experiment parameters\n",
    "num_iterations = 20\n",
    "num_rounds = 20\n",
    "num_epochs = 1\n",
    "labels_shared = 0\n",
    "test_size = 0.2\n",
    "\n",
    "\n",
    "\n",
    "all_experiment_results = []\n",
    "\n",
    "for iteration in range(num_iterations):\n",
    "    print(f\"Running iteration {iteration + 1}...\")\n",
    "    \n",
    "    # Initialize containers to store results for this iteration\n",
    "    iteration_global_models = []\n",
    "    iteration_perf = []\n",
    "    iteration_label_names = []\n",
    "    iteration_metrics = []\n",
    "    \n",
    "    #Loadind up train test split\n",
    "    X_train, X_test, y_train, y_test = train_test_splits[iteration]\n",
    "    # Federated Learning Experiment for this iteration\n",
    "    for n_nodes in range(2, 11):\n",
    "        nodes = create_FL_Datasets(X_train, y_train, n_nodes, labels_shared)\n",
    "        global_model = create_mlp_model(allEmbeddings.shape[1], labels.shape[1])\n",
    "        max_label = labels.shape[1] - 1\n",
    "        \n",
    "        # Perform federated learning\n",
    "        for round in range(num_rounds):\n",
    "            for node in nodes:\n",
    "                node['classifier'].fit(node['X_train'], node['y_train'], epochs=num_epochs, verbose=0)\n",
    "            global_model, nodes = surgical_aggregation(nodes, max_label, global_model)\n",
    "        \n",
    "        # Evaluate the global model\n",
    "        perf = global_model.evaluate(X_test, y_test, verbose=0)\n",
    "        iteration_global_models.append(global_model)\n",
    "        iteration_perf.append(perf)\n",
    "        label_names = [node['label_names'].tolist() for node in nodes]\n",
    "        iteration_label_names.append(label_names)\n",
    "        metrics = get_metrics(global_model, X_test, y_test, labels)\n",
    "        iteration_metrics.append(metrics)\n",
    "    \n",
    "    # Convert data to JSON-serializable format\n",
    "    iteration_perf_serializable = make_serializable(iteration_perf)\n",
    "    iteration_label_names_serializable = make_serializable(iteration_label_names)\n",
    "    iteration_metrics_serializable = make_serializable(iteration_metrics)\n",
    "    \n",
    "    # Store results of this iteration in a specific folder\n",
    "    iteration_dir = os.path.join(output_dir, f'iteration_{iteration + 1}')\n",
    "    os.makedirs(iteration_dir, exist_ok=True)\n",
    "    \n",
    "    # Save performance metrics as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_perf.json'), 'w') as f:\n",
    "        json.dump(iteration_perf_serializable, f)\n",
    "    \n",
    "    # Save label names as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_label_names.json'), 'w') as f:\n",
    "        json.dump(iteration_label_names_serializable, f)\n",
    "    \n",
    "    # Save detailed metrics as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_metrics.json'), 'w') as f:\n",
    "        json.dump(iteration_metrics_serializable, f)\n",
    "    \n",
    "    # Save global models\n",
    "    models_dir = os.path.join(iteration_dir, 'models')\n",
    "    os.makedirs(models_dir, exist_ok=True)\n",
    "    for idx, model in enumerate(iteration_global_models):\n",
    "        # Save model architecture to JSON\n",
    "        model_json = model.to_json()\n",
    "        with open(os.path.join(models_dir, f'global_model_{idx}.json'), 'w') as json_file:\n",
    "            json_file.write(model_json)\n",
    "        # Save model weights (weights cannot be saved as JSON)\n",
    "        model.save_weights(os.path.join(models_dir, f'global_model_{idx}.weights.h5'))\n",
    "    \n",
    "    # Save iteration-level data for comparison\n",
    "    experiment_data = {\n",
    "        'perf': iteration_perf_serializable,\n",
    "        'label_names': iteration_label_names_serializable,\n",
    "        'metrics': iteration_metrics_serializable\n",
    "    }\n",
    "    all_experiment_results.append(experiment_data)\n",
    "\n",
    "# Save all experiment results as JSON\n",
    "all_experiment_results_serializable = make_serializable(all_experiment_results)\n",
    "with open(os.path.join(output_dir, 'all_experiment_results.json'), 'w') as f:\n",
    "    json.dump(all_experiment_results_serializable, f)\n",
    "\n",
    "print(\"All 20 iterations are complete, and results are saved.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparing Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.0877254456281662, 0.979181706905365], [0.10726558417081833, 0.9658258557319641], [0.1122417151927948, 0.9631524682044983], [0.13514091074466705, 0.9459066390991211], [0.13544024527072906, 0.9452823400497437], [0.1559024304151535, 0.9331587553024292], [0.15732593834400177, 0.9310252666473389], [0.17007902264595032, 0.9225221872329712], [0.18038468062877655, 0.9144441485404968]]\n"
     ]
    }
   ],
   "source": [
    "with open('data/Experiment 2/all_experiment_results.json', 'r') as f:\n",
    "    all_experiment_results = json.load(f)\n",
    "\n",
    "# Access specific iteration results, for example, iteration 1:\n",
    "iteration_1_results = all_experiment_results[0]  # Index starts at 0\n",
    "\n",
    "# Print performance metrics for iteration 1\n",
    "print(iteration_1_results['perf'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the directory where the experiment data is stored\n",
    "experiment_dir = 'data/Experiment 2'\n",
    "\n",
    "# Load the overall results of all experiments\n",
    "with open(os.path.join(experiment_dir, 'all_experiment_results.json'), 'r') as f:\n",
    "    all_experiment_results = json.load(f)\n",
    "\n",
    "\n",
    "all_auc = []      # To store AUC values across iterations\n",
    "all_accuracy = [] # To store accuracy values across iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "for experiment in all_experiment_results:\n",
    "    perf_metrics = experiment['perf']\n",
    "    auc_values = [perf[1] for perf in perf_metrics]  # Assuming AUC is at index 1 in 'perf'\n",
    "    # Extract 'Accuracy' from 'metrics' dictionary for each node\n",
    "    accuracy_values = [metric['Overall']['Accuracy'] for metric in experiment['metrics']]\n",
    "    \n",
    "    all_auc.append(auc_values)\n",
    "    all_accuracy.append(accuracy_values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert lists to NumPy arrays for easy calculation of mean and std dev\n",
    "all_auc = np.array(all_auc)\n",
    "all_accuracy = np.array(all_accuracy)\n",
    "\n",
    "# Calculate mean and standard deviation for AUC and Accuracy across all iterations\n",
    "mean_auc = np.mean(all_auc, axis=0)\n",
    "std_auc = np.std(all_auc, axis=0)\n",
    "\n",
    "mean_accuracy = np.mean(all_accuracy, axis=0)\n",
    "std_accuracy = np.std(all_accuracy, axis=0)\n",
    "\n",
    "# Define the number of nodes used during the experiment\n",
    "n_nodes = np.arange(2, 11)  # Node range from 2 to 10 inclusive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting Mean AUC with Standard Deviation\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.errorbar(n_nodes, mean_auc, yerr=std_auc, fmt='-o', capsize=5, label='AUC')\n",
    "# Base model is a constant baseline (no error bars needed):\n",
    "plt.plot(\n",
    "    n_nodes,\n",
    "    [base_perf[1]] * len(n_nodes),\n",
    "    'r--',\n",
    "    label='Base Model'\n",
    ")\n",
    "\n",
    "plt.xlabel('Number of Nodes')\n",
    "plt.ylabel('Mean AUC ± Std Dev')\n",
    "plt.title('Mean AUC vs Number of Nodes with Standard Deviation')\n",
    "plt.ylim(0.0, 1)\n",
    "plt.xticks(n_nodes)\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting Mean Accuracy with Standard Deviation\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.errorbar(n_nodes, mean_accuracy, yerr=std_accuracy, fmt='-o', capsize=5, label='Accuracy')\n",
    "\n",
    "# Base model is a constant baseline (no error bars needed):\n",
    "plt.plot(\n",
    "    n_nodes,\n",
    "    [base_perf[1]] * len(n_nodes),\n",
    "    'r--',\n",
    "    label='Base Model'\n",
    ")\n",
    "\n",
    "plt.xlabel('Number of Nodes')\n",
    "plt.ylabel('Mean Accuracy ± Std Dev')\n",
    "plt.title('Mean Accuracy vs Number of Nodes with Standard Deviation')\n",
    "plt.ylim(0.0, 1)\n",
    "plt.xticks(n_nodes)\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Experiment 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.models import model_from_json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Function to make data JSON serializable\n",
    "def make_serializable(obj):\n",
    "    if isinstance(obj, dict):\n",
    "        return {k: make_serializable(v) for k, v in obj.items()}\n",
    "    elif isinstance(obj, list):\n",
    "        return [make_serializable(v) for v in obj]\n",
    "    elif isinstance(obj, np.ndarray):\n",
    "        return obj.tolist()\n",
    "    elif isinstance(obj, np.integer):\n",
    "        return int(obj)\n",
    "    elif isinstance(obj, np.floating):\n",
    "        return float(obj)\n",
    "    else:\n",
    "        return obj"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "--- Iteration 1/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 2/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 3/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 4/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 5/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 6/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 7/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 8/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 9/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 10/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 11/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 12/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 13/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 14/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 15/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 16/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 17/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 18/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 19/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "\n",
      "--- Iteration 20/20 ---\n",
      "  Testing shared_labels=0...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=1...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=2...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "  Testing shared_labels=3...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=4...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=5...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=6...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=7...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=8...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=9...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=10...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=11...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=12...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n",
      "  Testing shared_labels=13...\n",
      "\u001b[1m23/23\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step \n",
      "\n",
      "Experiment complete! Results saved to: data\\Experiment 3\n"
     ]
    }
   ],
   "source": [
    "# Define the output directory for Experiment 3\n",
    "experiment_dir = os.path.join('data', 'Experiment 3')\n",
    "os.makedirs(experiment_dir, exist_ok=True)\n",
    "\n",
    "# Initialize a list to hold all experiment results\n",
    "all_experiment_results = []\n",
    "num_iterations = 20\n",
    "num_shared_labels_range = range(0, 14)  # 0-13\n",
    "n_nodes = 2\n",
    "\n",
    "for iteration in range(num_iterations):\n",
    "    print(f\"\\n--- Iteration {iteration+1}/{num_iterations} ---\")\n",
    "    \n",
    "    # Per-iteration storage\n",
    "    iteration_results = []\n",
    "    \n",
    "    #Loadind up train test split\n",
    "    X_train, X_test, y_train, y_test = train_test_splits[iteration]\n",
    "\n",
    "    # Create iteration directory\n",
    "    iteration_dir = os.path.join(experiment_dir, f'iteration_{iteration+1}')\n",
    "    os.makedirs(iteration_dir, exist_ok=True)\n",
    "\n",
    "    for shared_labels in num_shared_labels_range:\n",
    "        print(f\"  Testing shared_labels={shared_labels}...\")\n",
    "        \n",
    "        # Initialize nodes and model\n",
    "        nodes = create_FL_Datasets(X_train, y_train, n_nodes, shared_labels)\n",
    "        global_model = create_mlp_model(allEmbeddings.shape[1], labels.shape[1])\n",
    "        max_label = labels.shape[1] - 1\n",
    "\n",
    "        # Federated training\n",
    "        for round in range(num_rounds):\n",
    "            for node in nodes:\n",
    "                node['classifier'].fit(node['X_train'], node['y_train'], \n",
    "                                      epochs=num_epochs, verbose=0)\n",
    "            global_model, nodes = surgical_aggregation(nodes, max_label, global_model)\n",
    "\n",
    "        # Evaluation\n",
    "        perf = global_model.evaluate(X_test, y_test, verbose=0)\n",
    "        metrics = get_metrics(global_model, X_test, y_test, labels)\n",
    "        label_names = [node['label_names'].tolist() for node in nodes]\n",
    "\n",
    "        # Create result entry\n",
    "        result = {\n",
    "            'shared_labels': shared_labels,\n",
    "            'perf': make_serializable(perf),\n",
    "            'metrics': make_serializable(metrics),\n",
    "            'label_names': make_serializable(label_names)\n",
    "        }\n",
    "        \n",
    "        # Save per-configuration files\n",
    "        config_dir = os.path.join(iteration_dir, f'shared_labels_{shared_labels}')\n",
    "        os.makedirs(config_dir, exist_ok=True)\n",
    "        \n",
    "        # Save metrics\n",
    "        with open(os.path.join(config_dir, 'metrics.json'), 'w') as f:\n",
    "            json.dump(result, f)\n",
    "        \n",
    "        # Save model\n",
    "        model_path = os.path.join(config_dir, 'global_model')\n",
    "        global_model.save_weights(model_path+'.weights.h5')  # Saves both architecture and weights\n",
    "        \n",
    "        iteration_results.append(result)\n",
    "\n",
    "    # Save iteration-level aggregation\n",
    "    with open(os.path.join(iteration_dir, 'iteration_summary.json'), 'w') as f:\n",
    "        json.dump(make_serializable(iteration_results), f)\n",
    "    \n",
    "    all_experiment_results.extend(iteration_results)\n",
    "\n",
    "# Save global results\n",
    "with open(os.path.join(experiment_dir, 'all_experiment_results.json'), 'w') as f:\n",
    "    json.dump(make_serializable(all_experiment_results), f)\n",
    "\n",
    "print(\"\\nExperiment complete! Results saved to:\", experiment_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running iteration 1...\n",
      "  Running with 0 shared labels...\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[33], line 36\u001b[0m\n\u001b[0;32m     34\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m \u001b[38;5;28mround\u001b[39m \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(num_rounds):\n\u001b[0;32m     35\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m node \u001b[38;5;129;01min\u001b[39;00m nodes:\n\u001b[1;32m---> 36\u001b[0m         \u001b[43mnode\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mclassifier\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnode\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mX_train\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnode\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43my_train\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnum_epochs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mverbose\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[0;32m     37\u001b[0m     global_model, nodes \u001b[38;5;241m=\u001b[39m surgical_aggregation(nodes, max_label, global_model)\n\u001b[0;32m     39\u001b[0m perf \u001b[38;5;241m=\u001b[39m global_model\u001b[38;5;241m.\u001b[39mevaluate(X_test, y_test, verbose\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0\u001b[39m)\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\keras\\src\\utils\\traceback_utils.py:117\u001b[0m, in \u001b[0;36mfilter_traceback.<locals>.error_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m    115\u001b[0m filtered_tb \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m    116\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 117\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    118\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[0;32m    119\u001b[0m     filtered_tb \u001b[38;5;241m=\u001b[39m _process_traceback_frames(e\u001b[38;5;241m.\u001b[39m__traceback__)\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\keras\\src\\backend\\tensorflow\\trainer.py:320\u001b[0m, in \u001b[0;36mTensorFlowTrainer.fit\u001b[1;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq)\u001b[0m\n\u001b[0;32m    318\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m step, iterator \u001b[38;5;129;01min\u001b[39;00m epoch_iterator\u001b[38;5;241m.\u001b[39menumerate_epoch():\n\u001b[0;32m    319\u001b[0m     callbacks\u001b[38;5;241m.\u001b[39mon_train_batch_begin(step)\n\u001b[1;32m--> 320\u001b[0m     logs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain_function\u001b[49m\u001b[43m(\u001b[49m\u001b[43miterator\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    321\u001b[0m     callbacks\u001b[38;5;241m.\u001b[39mon_train_batch_end(step, logs)\n\u001b[0;32m    322\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstop_training:\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\util\\traceback_utils.py:150\u001b[0m, in \u001b[0;36mfilter_traceback.<locals>.error_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m    148\u001b[0m filtered_tb \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m    149\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 150\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    151\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[0;32m    152\u001b[0m   filtered_tb \u001b[38;5;241m=\u001b[39m _process_traceback_frames(e\u001b[38;5;241m.\u001b[39m__traceback__)\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\polymorphic_function.py:833\u001b[0m, in \u001b[0;36mFunction.__call__\u001b[1;34m(self, *args, **kwds)\u001b[0m\n\u001b[0;32m    830\u001b[0m compiler \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mxla\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_jit_compile \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnonXla\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    832\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m OptionalXlaContext(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_jit_compile):\n\u001b[1;32m--> 833\u001b[0m   result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    835\u001b[0m new_tracing_count \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexperimental_get_tracing_count()\n\u001b[0;32m    836\u001b[0m without_tracing \u001b[38;5;241m=\u001b[39m (tracing_count \u001b[38;5;241m==\u001b[39m new_tracing_count)\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\polymorphic_function.py:878\u001b[0m, in \u001b[0;36mFunction._call\u001b[1;34m(self, *args, **kwds)\u001b[0m\n\u001b[0;32m    875\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock\u001b[38;5;241m.\u001b[39mrelease()\n\u001b[0;32m    876\u001b[0m \u001b[38;5;66;03m# In this case we have not created variables on the first call. So we can\u001b[39;00m\n\u001b[0;32m    877\u001b[0m \u001b[38;5;66;03m# run the first trace but we should fail if variables are created.\u001b[39;00m\n\u001b[1;32m--> 878\u001b[0m results \u001b[38;5;241m=\u001b[39m \u001b[43mtracing_compilation\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcall_function\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m    879\u001b[0m \u001b[43m    \u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mkwds\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_variable_creation_config\u001b[49m\n\u001b[0;32m    880\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    881\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_created_variables:\n\u001b[0;32m    882\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mCreating variables on a non-first call to a function\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    883\u001b[0m                    \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m decorated with tf.function.\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\tracing_compilation.py:139\u001b[0m, in \u001b[0;36mcall_function\u001b[1;34m(args, kwargs, tracing_options)\u001b[0m\n\u001b[0;32m    137\u001b[0m bound_args \u001b[38;5;241m=\u001b[39m function\u001b[38;5;241m.\u001b[39mfunction_type\u001b[38;5;241m.\u001b[39mbind(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m    138\u001b[0m flat_inputs \u001b[38;5;241m=\u001b[39m function\u001b[38;5;241m.\u001b[39mfunction_type\u001b[38;5;241m.\u001b[39munpack_inputs(bound_args)\n\u001b[1;32m--> 139\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunction\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_flat\u001b[49m\u001b[43m(\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# pylint: disable=protected-access\u001b[39;49;00m\n\u001b[0;32m    140\u001b[0m \u001b[43m    \u001b[49m\u001b[43mflat_inputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcaptured_inputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mfunction\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcaptured_inputs\u001b[49m\n\u001b[0;32m    141\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\concrete_function.py:1322\u001b[0m, in \u001b[0;36mConcreteFunction._call_flat\u001b[1;34m(self, tensor_inputs, captured_inputs)\u001b[0m\n\u001b[0;32m   1318\u001b[0m possible_gradient_type \u001b[38;5;241m=\u001b[39m gradients_util\u001b[38;5;241m.\u001b[39mPossibleTapeGradientTypes(args)\n\u001b[0;32m   1319\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (possible_gradient_type \u001b[38;5;241m==\u001b[39m gradients_util\u001b[38;5;241m.\u001b[39mPOSSIBLE_GRADIENT_TYPES_NONE\n\u001b[0;32m   1320\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m executing_eagerly):\n\u001b[0;32m   1321\u001b[0m   \u001b[38;5;66;03m# No tape is watching; skip to running the function.\u001b[39;00m\n\u001b[1;32m-> 1322\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_inference_function\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcall_preflattened\u001b[49m\u001b[43m(\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   1323\u001b[0m forward_backward \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_select_forward_and_backward_functions(\n\u001b[0;32m   1324\u001b[0m     args,\n\u001b[0;32m   1325\u001b[0m     possible_gradient_type,\n\u001b[0;32m   1326\u001b[0m     executing_eagerly)\n\u001b[0;32m   1327\u001b[0m forward_function, args_with_tangents \u001b[38;5;241m=\u001b[39m forward_backward\u001b[38;5;241m.\u001b[39mforward()\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\atomic_function.py:216\u001b[0m, in \u001b[0;36mAtomicFunction.call_preflattened\u001b[1;34m(self, args)\u001b[0m\n\u001b[0;32m    214\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mcall_preflattened\u001b[39m(\u001b[38;5;28mself\u001b[39m, args: Sequence[core\u001b[38;5;241m.\u001b[39mTensor]) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Any:\n\u001b[0;32m    215\u001b[0m \u001b[38;5;250m  \u001b[39m\u001b[38;5;124;03m\"\"\"Calls with flattened tensor inputs and returns the structured output.\"\"\"\u001b[39;00m\n\u001b[1;32m--> 216\u001b[0m   flat_outputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcall_flat\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    217\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mfunction_type\u001b[38;5;241m.\u001b[39mpack_output(flat_outputs)\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\atomic_function.py:251\u001b[0m, in \u001b[0;36mAtomicFunction.call_flat\u001b[1;34m(self, *args)\u001b[0m\n\u001b[0;32m    249\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m record\u001b[38;5;241m.\u001b[39mstop_recording():\n\u001b[0;32m    250\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_bound_context\u001b[38;5;241m.\u001b[39mexecuting_eagerly():\n\u001b[1;32m--> 251\u001b[0m     outputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_bound_context\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcall_function\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m    252\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mname\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    253\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mlist\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    254\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfunction_type\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mflat_outputs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    255\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    256\u001b[0m   \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m    257\u001b[0m     outputs \u001b[38;5;241m=\u001b[39m make_call_op_in_graph(\n\u001b[0;32m    258\u001b[0m         \u001b[38;5;28mself\u001b[39m,\n\u001b[0;32m    259\u001b[0m         \u001b[38;5;28mlist\u001b[39m(args),\n\u001b[0;32m    260\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_bound_context\u001b[38;5;241m.\u001b[39mfunction_call_options\u001b[38;5;241m.\u001b[39mas_attrs(),\n\u001b[0;32m    261\u001b[0m     )\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\context.py:1552\u001b[0m, in \u001b[0;36mContext.call_function\u001b[1;34m(self, name, tensor_inputs, num_outputs)\u001b[0m\n\u001b[0;32m   1550\u001b[0m cancellation_context \u001b[38;5;241m=\u001b[39m cancellation\u001b[38;5;241m.\u001b[39mcontext()\n\u001b[0;32m   1551\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m cancellation_context \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m-> 1552\u001b[0m   outputs \u001b[38;5;241m=\u001b[39m \u001b[43mexecute\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mexecute\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m   1553\u001b[0m \u001b[43m      \u001b[49m\u001b[43mname\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdecode\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mutf-8\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   1554\u001b[0m \u001b[43m      \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnum_outputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   1555\u001b[0m \u001b[43m      \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtensor_inputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   1556\u001b[0m \u001b[43m      \u001b[49m\u001b[43mattrs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m   1557\u001b[0m \u001b[43m      \u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[0;32m   1558\u001b[0m \u001b[43m  \u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   1559\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m   1560\u001b[0m   outputs \u001b[38;5;241m=\u001b[39m execute\u001b[38;5;241m.\u001b[39mexecute_with_cancellation(\n\u001b[0;32m   1561\u001b[0m       name\u001b[38;5;241m.\u001b[39mdecode(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mutf-8\u001b[39m\u001b[38;5;124m\"\u001b[39m),\n\u001b[0;32m   1562\u001b[0m       num_outputs\u001b[38;5;241m=\u001b[39mnum_outputs,\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m   1566\u001b[0m       cancellation_manager\u001b[38;5;241m=\u001b[39mcancellation_context,\n\u001b[0;32m   1567\u001b[0m   )\n",
      "File \u001b[1;32mc:\\Users\\nikhi\\Desktop\\GITHUB PROJECTS\\Federated-MedEmbedX\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\execute.py:53\u001b[0m, in \u001b[0;36mquick_execute\u001b[1;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[0;32m     51\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m     52\u001b[0m   ctx\u001b[38;5;241m.\u001b[39mensure_initialized()\n\u001b[1;32m---> 53\u001b[0m   tensors \u001b[38;5;241m=\u001b[39m \u001b[43mpywrap_tfe\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mTFE_Py_Execute\u001b[49m\u001b[43m(\u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_handle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdevice_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mop_name\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m     54\u001b[0m \u001b[43m                                      \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     55\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m core\u001b[38;5;241m.\u001b[39m_NotOkStatusException \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[0;32m     56\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "\n",
    "#OLD VERSION\n",
    "\n",
    "# Define the output directory for Experiment 3\n",
    "experiment_dir = os.path.join('data', 'Experiment 3')\n",
    "if not os.path.exists(experiment_dir):\n",
    "    os.makedirs(experiment_dir)\n",
    "\n",
    "# Initialize a list to hold all experiment results\n",
    "all_experiment_results = []\n",
    "num_iterations = 20\n",
    "num_shared_labels_range = range(0, 14)  # From 0 to 13\n",
    "n_nodes = 2  # Number of nodes fixed at 2\n",
    "test_size = 0.2\n",
    "num_rounds = 20\n",
    "num_epochs = 1\n",
    "\n",
    "for iteration in range(num_iterations):\n",
    "    print(f\"Running iteration {iteration + 1}...\")\n",
    "    \n",
    "    # Initialize containers to store results for this iteration\n",
    "    iteration_global_models = []\n",
    "    iteration_perf = []\n",
    "    iteration_label_names = []\n",
    "    iteration_metrics = []\n",
    "    \n",
    "    # Loop over shared labels\n",
    "    for shared_labels in num_shared_labels_range:\n",
    "        print(f\"  Running with {shared_labels} shared labels...\")\n",
    "        \n",
    "        # Initialize nodes and model\n",
    "        nodes = create_FL_Datasets(X_train, y_train, n_nodes, shared_labels)\n",
    "        global_model = create_mlp_model(allEmbeddings.shape[1], labels.shape[1])\n",
    "        max_label = labels.shape[1] - 1\n",
    "        \n",
    "        # Perform federated learning\n",
    "        for round in range(num_rounds):\n",
    "            for node in nodes:\n",
    "                node['classifier'].fit(node['X_train'], node['y_train'], epochs=num_epochs, verbose=0)\n",
    "            global_model, nodes = surgical_aggregation(nodes, max_label, global_model)\n",
    "        \n",
    "        perf = global_model.evaluate(X_test, y_test, verbose=0)\n",
    "        iteration_global_models.append(global_model)\n",
    "        iteration_perf.append(perf)\n",
    "        label_names = [node['label_names'].tolist() for node in nodes]\n",
    "        iteration_label_names.append(label_names)\n",
    "        metrics = get_metrics(global_model, X_test, y_test, labels)\n",
    "        iteration_metrics.append(metrics)\n",
    "        \n",
    "    # Convert data to JSON-serializable format\n",
    "    iteration_perf_serializable = make_serializable(iteration_perf)\n",
    "    iteration_label_names_serializable = make_serializable(iteration_label_names)\n",
    "    iteration_metrics_serializable = make_serializable(iteration_metrics)\n",
    "        \n",
    "    # Store results of this iteration in a specific folder\n",
    "    iteration_dir = os.path.join(output_dir, f'iteration_{iteration + 1}')\n",
    "    os.makedirs(iteration_dir, exist_ok=True)\n",
    "    \n",
    "    # Save performance metrics as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_perf.json'), 'w') as f:\n",
    "        json.dump(iteration_perf_serializable, f)\n",
    "    \n",
    "    # Save label names as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_label_names.json'), 'w') as f:\n",
    "        json.dump(iteration_label_names_serializable, f)\n",
    "    \n",
    "    # Save detailed metrics as JSON\n",
    "    with open(os.path.join(iteration_dir, 'iteration_metrics.json'), 'w') as f:\n",
    "        json.dump(iteration_metrics_serializable, f)\n",
    "    \n",
    "    # Save global models\n",
    "    models_dir = os.path.join(iteration_dir, 'models')\n",
    "    os.makedirs(models_dir, exist_ok=True)\n",
    "    for idx, model in enumerate(iteration_global_models):\n",
    "        # Save model architecture to JSON\n",
    "        model_json = model.to_json()\n",
    "        with open(os.path.join(models_dir, f'global_model_{idx}.json'), 'w') as json_file:\n",
    "            json_file.write(model_json)\n",
    "        # Save model weights (weights cannot be saved as JSON)\n",
    "        model.save_weights(os.path.join(models_dir, f'global_model_{idx}.weights.h5'))\n",
    "    \n",
    "    # Save iteration-level data for comparison\n",
    "    experiment_data = {\n",
    "        'perf': iteration_perf_serializable,\n",
    "        'label_names': iteration_label_names_serializable,\n",
    "        'metrics': iteration_metrics_serializable\n",
    "    }\n",
    "    all_experiment_results.append(experiment_data)\n",
    "\n",
    "# Save all experiment results as JSON\n",
    "all_experiment_results_serializable = make_serializable(all_experiment_results)\n",
    "with open(os.path.join(output_dir, 'all_experiment_results.json'), 'w') as f:\n",
    "    json.dump(all_experiment_results_serializable, f)\n",
    "\n",
    "print(\"All 20 iterations are complete, and results are saved.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparing Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.08726181089878082, 0.9797136187553406]\n"
     ]
    }
   ],
   "source": [
    "with open('data/Experiment 3/all_experiment_results.json', 'r') as f:\n",
    "    all_experiment_results = json.load(f)\n",
    "\n",
    "# Access specific iteration results, for example, iteration 1:\n",
    "iteration_1_results = all_experiment_results[0]  # Index starts at 0\n",
    "\n",
    "# Print performance metrics for iteration 1\n",
    "print(iteration_1_results['perf'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualisation New"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Initialize containers for metrics\n",
    "shared_labels_list = list(range(0, 14))\n",
    "mean_auc_list = []\n",
    "std_auc_list = []\n",
    "mean_accuracy_list = []\n",
    "std_accuracy_list = []\n",
    "\n",
    "# For each possible shared_labels value, collect metrics across \"iterations\" (i.e., across experiments)\n",
    "for shared_labels in shared_labels_list:\n",
    "    auc_values = []\n",
    "    accuracy_values = []\n",
    "    \n",
    "    # Loop over all experiments in all_experiment_results\n",
    "    for iteration_data in all_experiment_results:\n",
    "        if iteration_data['shared_labels'] == shared_labels:\n",
    "            # 'perf' is a list like [threshold, AUC]\n",
    "            perf = iteration_data['perf']\n",
    "            # 'metrics' is a dictionary with 'Overall' among other keys\n",
    "            metrics = iteration_data['metrics']\n",
    "            \n",
    "            # perf[1] is AUC\n",
    "            auc_values.append(perf[1])\n",
    "            \n",
    "            # Overall accuracy\n",
    "            accuracy_values.append(metrics['Overall']['Accuracy'])\n",
    "    \n",
    "    # If no experiment matched this shared_labels (i.e., empty list), handle carefully:\n",
    "    if len(auc_values) == 0:\n",
    "        mean_auc_list.append(np.nan)  # or 0, or skip altogether\n",
    "        std_auc_list.append(np.nan)\n",
    "        mean_accuracy_list.append(np.nan)\n",
    "        std_accuracy_list.append(np.nan)\n",
    "    else:\n",
    "        # Calculate mean and std dev for this shared_labels value\n",
    "        mean_auc_list.append(np.mean(auc_values))\n",
    "        std_auc_list.append(np.std(auc_values))\n",
    "        mean_accuracy_list.append(np.mean(accuracy_values))\n",
    "        std_accuracy_list.append(np.std(accuracy_values))\n",
    "\n",
    "# Convert lists to NumPy arrays\n",
    "mean_auc_array = np.array(mean_auc_list)\n",
    "std_auc_array = np.array(std_auc_list)\n",
    "mean_accuracy_array = np.array(mean_accuracy_list)\n",
    "std_accuracy_array = np.array(std_accuracy_list)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Plotting Mean AUC with Standard Deviation\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.errorbar(shared_labels_list, mean_auc_array, yerr=std_auc_array, fmt='-o', capsize=5)\n",
    "plt.xlabel('Number of Shared Labels')\n",
    "plt.ylabel('Mean AUC ± Std Dev')\n",
    "plt.title('Mean AUC vs Number of Shared Labels')\n",
    "plt.plot(\n",
    "    shared_labels_list,\n",
    "    [base_perf[1]] * len(shared_labels_list),  # base_perf[1] is your reference AUC\n",
    "    'r--',\n",
    "    label='Base Model'\n",
    ")\n",
    "plt.ylim(0.0, 1)\n",
    "plt.xticks(shared_labels_list)\n",
    "plt.grid(True)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Plotting Mean Accuracy with Standard Deviation\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.errorbar(shared_labels_list, mean_accuracy_array, yerr=std_accuracy_array, fmt='-o', capsize=5)\n",
    "plt.xlabel('Number of Shared Labels')\n",
    "plt.ylabel('Mean Accuracy ± Std Dev')\n",
    "plt.title('Mean Accuracy vs Number of Shared Labels')\n",
    "plt.plot(\n",
    "    shared_labels_list,\n",
    "    [base_perf[1]] * len(shared_labels_list),  # If you also want to compare accuracy to the same base_perf[1] \n",
    "    'r--',\n",
    "    label='Base Model'\n",
    ")\n",
    "plt.xticks(shared_labels_list)\n",
    "plt.ylim(0.0, 1)\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# T-Tests"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import ttest_rel\n",
    "def perform_t_tests(results, label_columns, metrics=['Accuracy', 'Sensitivity', 'Specificity', 'AUC']):\n",
    "    t_test_results = {}\n",
    "\n",
    "    configs = list(results.keys())\n",
    "    for i in range(len(configs)):\n",
    "        for j in range(i + 1, len(configs)):\n",
    "            config_1 = configs[i]\n",
    "            config_2 = configs[j]\n",
    "            comparison_key = f'{config_1} vs {config_2}'\n",
    "            t_test_results[comparison_key] = {}\n",
    "\n",
    "            for metric in metrics:\n",
    "                data_config_1 = results[config_1][metric]\n",
    "                data_config_2 = results[config_2][metric]\n",
    "\n",
    "                # Perform the paired t-test\n",
    "                t_stat, p_value = ttest_rel(data_config_1, data_config_2)\n",
    "                t_test_results[comparison_key][metric] = {'t_stat': t_stat, 'p_value': p_value}\n",
    "\n",
    "    return t_test_results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Experiment 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Two Nodes': {'Accuracy': [0.911848042816665, 0.9109035575611291, 0.9048168748032324, 0.9165704690943435, 0.903137789904502, 0.8966313359219225, 0.9006191625564066, 0.855178927484521, 0.908594815825375, 0.8888655682652954, 0.8712351768286285, 0.9195088676671214, 0.9190890964424389, 0.897365935565117, 0.893273166124462, 0.896736278728093, 0.8899149963270019, 0.899674677300871, 0.8945324797985099, 0.9169902403190261], 'Sensitivity': [0.9510768665174978, 0.9248276038282953, 0.9339544240742723, 0.9109686156539939, 0.9503786316367955, 0.9538067043948496, 0.9308939728790919, 0.9350199420821675, 0.9012726353392964, 0.9198857155479525, 0.9511727719057591, 0.9453204432441654, 0.9136022191646889, 0.9526941152624213, 0.9312858384133794, 0.9491030012583452, 0.9178956424297886, 0.9309900860324489, 0.9443416116084731, 0.9420822554602549], 'Specificity': [0.908441924046181, 0.9087967956472914, 0.902089113491974, 0.9168972709022118, 0.9032235414623769, 0.8938104328560007, 0.8979364516176993, 0.85358481210797, 0.9053408516854439, 0.890579549563819, 0.8688368258880922, 0.9203231320055385, 0.9200046436174285, 0.8958562514815342, 0.8923862525907157, 0.8955385993063433, 0.8866360430203931, 0.8987331513962526, 0.8955443483285849, 0.9156414860404146], 'AUC': [0.9671626533508336, 0.9561393405055953, 0.9568539065997328, 0.954194604296669, 0.9626118598038816, 0.9603036650465859, 0.9565719384396995, 0.9321864993146209, 0.9389694822020801, 0.9463657265666412, 0.950536073411217, 0.9664428296274976, 0.9534396616229995, 0.9600703805010038, 0.949854058509846, 0.9594438031292094, 0.9521394931828, 0.9579428805896123, 0.9608413246602312, 0.9630449349017787]}, 'Three Nodes': {'Accuracy': [0.8859271696925176, 0.8835134851505929, 0.8680868926435091, 0.8806800293839857, 0.8765872599433309, 0.887291426172736, 0.8822541714765453, 0.8837233707629342, 0.8869765977542239, 0.8670374645818029, 0.8887606254591248, 0.8877111973974183, 0.9000944485255536, 0.901458705005772, 0.8953720222478749, 0.8675621786126562, 0.8937978801553155, 0.90282296148599, 0.8861370553048589, 0.8949522510231922], 'Sensitivity': [0.9304545137372681, 0.9143355216898772, 0.9218698338653082, 0.9324972337941912, 0.9466172954435651, 0.9432421214558916, 0.9332319397006442, 0.9323651743029842, 0.8753640524066667, 0.9148773561535009, 0.9190424464203738, 0.9411634333482747, 0.8983187122533637, 0.9394032579968733, 0.8897625403775365, 0.9269783233468056, 0.8927097873115402, 0.9039448917845638, 0.9148780630334284, 0.9360895442103442], 'Specificity': [0.8853511566537003, 0.8826767987856806, 0.8680746408840553, 0.8780675019224888, 0.8757665195115725, 0.88536725076441, 0.882627246454175, 0.8838332759239059, 0.8899773911973182, 0.8645510073837208, 0.8844808288431706, 0.8883543679680534, 0.9008587806130959, 0.9017866341594197, 0.8974130214109586, 0.8656305543024845, 0.8931125685221273, 0.9012154379487213, 0.8895115126602265, 0.8934856803865187], 'AUC': [0.9466590672484225, 0.9415567140934221, 0.9418499459857878, 0.9486045208941317, 0.9424313440751344, 0.9511913943150238, 0.9502104126299276, 0.9410872802453325, 0.9313118087534585, 0.9377132594789424, 0.9393657295768106, 0.9503955271502298, 0.9383490147693597, 0.9558450678810863, 0.9291179603171534, 0.936553056459833, 0.9343983036338095, 0.9409345977422355, 0.9459389455407238, 0.948767756675341]}, 'Four Nodes': {'Accuracy': [0.8873963689789064, 0.8615804386609298, 0.8665127505509496, 0.8793157729037675, 0.8672473501941442, 0.8528701857487669, 0.8526603001364256, 0.8306223108405918, 0.8672473501941441, 0.8514009864623779, 0.861475495854759, 0.8573827264141041, 0.8494070731451361, 0.8831986567320809, 0.8465736173785287, 0.8396473921712667, 0.8465736173785287, 0.8877111973974183, 0.8409067058453144, 0.8633644663658306], 'Sensitivity': [0.8948213770894877, 0.8853296977005287, 0.9149623256170517, 0.9092383747721331, 0.9207645480307011, 0.9318972705383614, 0.9341093200465475, 0.9421641897803457, 0.9262606741404084, 0.939011691777585, 0.9034748132875617, 0.9073040111835999, 0.9284461065925012, 0.8955473902626621, 0.9096676997522036, 0.9107794882954932, 0.903259608566787, 0.8489402648480786, 0.918488250289835, 0.9455762773020285], 'Specificity': [0.8890490322687495, 0.860265774244296, 0.8634643538041692, 0.8796720079995938, 0.8672438892387146, 0.8494746941969592, 0.846436087888767, 0.8288000028985034, 0.8637173845681931, 0.8471010358728086, 0.8616044852582674, 0.8580056536831636, 0.8466499022772702, 0.8845666243884215, 0.8464013170323674, 0.8341219525977824, 0.8436083646284565, 0.8916948000590434, 0.8390697678802883, 0.8597225791447504], 'AUC': [0.9320451806668509, 0.911145512796748, 0.9287592158126035, 0.9326089604694536, 0.9276864228093735, 0.9288805380761916, 0.9320393989770762, 0.915964070757956, 0.9278324538887587, 0.9235498792851813, 0.9295062250825027, 0.9218703300442977, 0.9282076950969524, 0.93334232230865, 0.919766549181554, 0.9209466149316939, 0.9135664845811688, 0.9072600247391681, 0.9255282006718117, 0.9326268024026507]}, 'Five Nodes': {'Accuracy': [0.8212824010914053, 0.8455241893168224, 0.8319865673208101, 0.8097386924126351, 0.8336656522195405, 0.8424808479378739, 0.8334557666071991, 0.8342953090565642, 0.8509812152376953, 0.832301395739322, 0.8221219435407702, 0.8270542554307903, 0.8178192884877742, 0.8401721062021198, 0.801972924756008, 0.8231713716024767, 0.8044915521041033, 0.8277888550739847, 0.829992654003568, 0.8335607094133697], 'Sensitivity': [0.8905445466201625, 0.8889861387819684, 0.8998627554905827, 0.9098582742249314, 0.9101489815717281, 0.895009904496193, 0.8735460185674414, 0.9121313662208028, 0.8931320638637144, 0.9037269886534871, 0.8977541053941136, 0.9289830813579095, 0.9046231278427099, 0.9165541695740143, 0.924486173808707, 0.887784667760569, 0.9085054901061024, 0.881954453023683, 0.8918695643003635, 0.9219107073016727], 'Specificity': [0.8208969272282454, 0.8427277201204564, 0.8296598211944294, 0.804421019746595, 0.8333014730479218, 0.8433848619365515, 0.833376610535658, 0.8294994563330073, 0.8481046878429426, 0.8276422606927363, 0.8211411144399359, 0.8265514072725664, 0.815877756191828, 0.8388155300488139, 0.794765246837668, 0.823427776899252, 0.7993998775587579, 0.8271393106971845, 0.8298505970418659, 0.8279115901871832], 'AUC': [0.8914377269652435, 0.8955860368103653, 0.8944927333575341, 0.9056227306017461, 0.9000223072545184, 0.9006038704173991, 0.8984428565764907, 0.9104983234543458, 0.9137951949098835, 0.9037348354830167, 0.9009773944049287, 0.9210757153107748, 0.9033533264033208, 0.9209872531601486, 0.8959519818567875, 0.8937919240863266, 0.9028604973125366, 0.8922938758672498, 0.9073380133848168, 0.9058471915847732]}, 'Six Nodes': {'Accuracy': [0.8268443698184489, 0.7863364466365831, 0.8017630391436668, 0.7900094448525554, 0.8205478014482106, 0.8430055619687269, 0.79074404449575, 0.8231713716024766, 0.7982999265400357, 0.7869661034736067, 0.7750026235701543, 0.8236960856333297, 0.8175044600692623, 0.8404869346206317, 0.7839227620946585, 0.8036520096547383, 0.7889600167908488, 0.8053310945534685, 0.8372337076293419, 0.811207891699024], 'Sensitivity': [0.887835507388381, 0.8703979038067264, 0.9024762822756214, 0.9081339530582506, 0.9067507788648884, 0.9012634447466215, 0.9064975336792018, 0.8793584376272684, 0.9237622548668286, 0.9002898721268271, 0.9076688502413399, 0.897855080656291, 0.8869086486898863, 0.8804838336402903, 0.8975718587346534, 0.8811137234221899, 0.8499901048256592, 0.8932993473157613, 0.8442045006974475, 0.9179570338832094], 'Specificity': [0.824437109315676, 0.7852824410563211, 0.7975291679405084, 0.7849598077462578, 0.8177611810760481, 0.841763428956551, 0.7936962990864641, 0.8218350517206683, 0.793492433700476, 0.7808536040029914, 0.7702364092770414, 0.8212473866968251, 0.8150320753315277, 0.8399486568796977, 0.7811672374126916, 0.8033540633712902, 0.7895053932651214, 0.8044489649297699, 0.8384717399448485, 0.8067865120885436], 'AUC': [0.894381916406578, 0.8680204653118582, 0.8862597740197246, 0.8925692074146861, 0.8979267111510331, 0.9060761549434161, 0.8814973312708503, 0.8869085831120247, 0.898752487652692, 0.8843861931978665, 0.8740304189159762, 0.8977451146803409, 0.8992792024547416, 0.904826384834365, 0.8699505802814292, 0.8833881917052762, 0.8656443962348492, 0.8904193333935346, 0.8848653620077636, 0.8921981700718498]}, 'Seven Nodes': {'Accuracy': [0.8011333823066429, 0.7428901248819393, 0.7757372232133488, 0.7806695351033687, 0.7618847727988247, 0.7927379578129918, 0.7882254171476545, 0.7826634484206108, 0.7857067897995594, 0.7946269283240633, 0.7627243152481896, 0.7918984153636269, 0.7957812991919404, 0.8223318291531114, 0.7596809738692414, 0.7492916360583481, 0.8038618952670794, 0.8038618952670794, 0.7743729667331305, 0.779725049847833], 'Sensitivity': [0.8882941235641202, 0.8859588505711634, 0.9224882413312723, 0.8468226984412557, 0.912167141425409, 0.9043154901140434, 0.8757611854200615, 0.9201745761818193, 0.8856743144328909, 0.9007036028891513, 0.9014379197385552, 0.9073632193730906, 0.8543010388712917, 0.8659032026662167, 0.9001559509110837, 0.9090002651484888, 0.8328973629973478, 0.8697849644443829, 0.8701780208602357, 0.9215374584290411], 'Specificity': [0.797798071073098, 0.7350522959167562, 0.7668499887394085, 0.7739554551775704, 0.7561815340588566, 0.7892165000796244, 0.7846340625422784, 0.7786408849943138, 0.784475202236144, 0.7917326841593197, 0.7566315054991442, 0.7896660043298381, 0.7903468619704532, 0.8199918083413879, 0.7526191896206704, 0.7413948030269224, 0.8012987848962838, 0.7992180050261894, 0.7714693552445856, 0.7736751280412838], 'AUC': [0.8746060466513458, 0.8485938841511698, 0.8780518342718835, 0.8559938777793172, 0.8629306328828439, 0.8750022643126113, 0.8746751471278661, 0.882884364403295, 0.8788976152649177, 0.8869781604206745, 0.8626422560426824, 0.8818646764072583, 0.8726036924943, 0.8847305724191721, 0.862552341027457, 0.8641290589645322, 0.8595034635042283, 0.8732835954200174, 0.8678495147114921, 0.8759803984213914]}, 'Eight Nodes': {'Accuracy': [0.7751075663763248, 0.7630391436667018, 0.7949417567425754, 0.7069996851715814, 0.7294574456920979, 0.8198132018050164, 0.7480323223843006, 0.7574771749396578, 0.7471927799349355, 0.803127295623885, 0.7617798299926541, 0.7984048693462061, 0.7780459649491027, 0.8201280302235282, 0.7550634903977331, 0.7488718648336656, 0.7741630811207891, 0.7576870605519991, 0.7348095288068003, 0.798509812152377], 'Sensitivity': [0.8961237444005223, 0.8569655681024242, 0.8751664592211595, 0.8958657744269262, 0.9260159634542784, 0.8888251242541954, 0.8962581906845994, 0.895641788104265, 0.9297021444509519, 0.9036871653279067, 0.876949056913673, 0.8953840086386564, 0.8841383883142979, 0.8538253705207978, 0.8951809102747004, 0.8847761841235481, 0.8783102390633426, 0.8945630626245171, 0.8738283330970693, 0.8726176307585043], 'Specificity': [0.767716317236509, 0.7603071744795844, 0.7948920364894231, 0.6987492223379941, 0.7200312126694157, 0.8148910504524391, 0.7375183105412648, 0.7519823277728823, 0.7372383658392239, 0.7967587928692705, 0.7578078722100978, 0.7939827197135886, 0.7752145248739131, 0.8185468691736452, 0.742993051430546, 0.746561017363701, 0.7682300571348003, 0.7503475527790118, 0.7288263443624832, 0.7933557743411227], 'AUC': [0.8741209854585565, 0.8456620585852677, 0.8695706587742417, 0.845398526492533, 0.8607003837956881, 0.8845733061405283, 0.8579015297054036, 0.8628198821650249, 0.8699806845973429, 0.8809423639863615, 0.8488343051593318, 0.8832100143298475, 0.8680637505705493, 0.8765494662746117, 0.8520869176046174, 0.8636614910603536, 0.8539034149214919, 0.8619572723383006, 0.8395515735338722, 0.8625154679655679]}, 'Nine Nodes': {'Accuracy': [0.7773113653059082, 0.7416308112078916, 0.7727988246405708, 0.7619897156049953, 0.7440444957498165, 0.7659775422394796, 0.7269388183440026, 0.7636688005037257, 0.7881204743414839, 0.758631545807535, 0.755693147234757, 0.7744779095393011, 0.7801448210725155, 0.7722741106097176, 0.7406863259523558, 0.7824535628082696, 0.7810893063280513, 0.7786756217861265, 0.7686011123937454, 0.7200125931367404], 'Sensitivity': [0.894696519497944, 0.8806702749342481, 0.8947625134522809, 0.8408772691313368, 0.9420388028616783, 0.9135422179586004, 0.8701419368585575, 0.886895253444581, 0.9082142021117298, 0.9022745845373047, 0.8918736745089713, 0.9019765663487331, 0.8599860531451531, 0.8626446225094485, 0.8604345475398862, 0.8601561446001151, 0.8444993263597778, 0.8523066932316101, 0.8493593571322877, 0.9157206236873664], 'Specificity': [0.7680232135818295, 0.7377865392021944, 0.7690216761485291, 0.7528793407512924, 0.7293534892513877, 0.7608593682511001, 0.7193655099540854, 0.75753583617246, 0.777166312320542, 0.7532205180063648, 0.7484955954403657, 0.7680572487344522, 0.7730597334001933, 0.7688442984618943, 0.7341859862640601, 0.7769104567687795, 0.7795456358367857, 0.7777660748436067, 0.7636847357062154, 0.711459224088578], 'AUC': [0.8656448352367762, 0.8469451023806702, 0.8709177642076801, 0.8420885174148302, 0.8613339077105135, 0.8748842243544577, 0.8370037942112218, 0.858593512874616, 0.8800626582777406, 0.8657439631954822, 0.846427726347766, 0.8632691910602711, 0.8573896548209515, 0.8543876150136528, 0.8310636025252105, 0.8600780252611604, 0.8403943968942087, 0.8492369030002066, 0.8404210401745209, 0.8423445121980128]}, 'Ten Nodes': {'Accuracy': [0.7312414733969986, 0.7497114072830307, 0.7483471508028126, 0.7224262776786651, 0.7323958442648755, 0.8090040927694406, 0.7377479273795781, 0.7215867352293, 0.7506558925385665, 0.7578969461643404, 0.7392171266659672, 0.7590513170322175, 0.7774163081120791, 0.7801448210725154, 0.7237905341588834, 0.735754014062336, 0.7375380417672368, 0.7663973134641621, 0.7183335082380103, 0.7479273795781298], 'Sensitivity': [0.9223550853328779, 0.8607205232873038, 0.9003090891928113, 0.884727725195295, 0.9141589402847013, 0.8763594006177973, 0.8797991732823971, 0.926361066526058, 0.9059789940097023, 0.8815858945489083, 0.8695028727478497, 0.9022850737159458, 0.8256261607145959, 0.8645070857250897, 0.867321401306172, 0.8618586317542982, 0.8670110207464721, 0.8498589097826724, 0.8617121516420715, 0.9068892122660057], 'Specificity': [0.7215848844892523, 0.7465660166594333, 0.7347891701645172, 0.7086045285938529, 0.7170515714238537, 0.80519415808348, 0.7307968525389439, 0.7111058258077272, 0.7437901543169705, 0.7539224596087747, 0.7344015723722099, 0.7496533483291942, 0.7694728478617462, 0.7724900440974635, 0.7150883948133843, 0.730568482812671, 0.7284297353004755, 0.7613983658843052, 0.7092735376784649, 0.7423502946291698], 'AUC': [0.8510539515647058, 0.8344412916064733, 0.8549151109770107, 0.8393970170133012, 0.8390453740376675, 0.8693217751000704, 0.8455010170971777, 0.8481620249533807, 0.8659244245246385, 0.856485134939613, 0.8345143758582949, 0.8526664974699083, 0.8356662102052632, 0.8568615979912918, 0.8287370849285646, 0.8332358216497757, 0.8321544705911867, 0.8329382757325305, 0.8211825487580563, 0.8452590486197932]}}\n"
     ]
    }
   ],
   "source": [
    "import glob\n",
    "import os\n",
    "import json\n",
    "experiment_dir='data\\Experiment 2'\n",
    "results={}\n",
    "metrics=['Accuracy', 'Sensitivity', 'Specificity', 'AUC']\n",
    "label_list= ['Two Nodes', 'Three Nodes', 'Four Nodes', 'Five Nodes', 'Six Nodes', 'Seven Nodes', 'Eight Nodes', 'Nine Nodes', 'Ten Nodes']\n",
    "results = {\n",
    "    label: {\n",
    "        \"Accuracy\": [],\n",
    "        \"Sensitivity\": [],\n",
    "        \"Specificity\": [],\n",
    "        \"AUC\": []\n",
    "    } for label in label_list\n",
    "}\n",
    "for count in range(1,21):\n",
    "    with open(experiment_dir+'\\iteration_'+str(count)+'\\iteration_metrics.json', 'r') as f:\n",
    "        data = json.load(f)\n",
    "    for item,label_name in zip(data,label_list):\n",
    "        for metrics, value in item['Overall'].items():\n",
    "            results[label_name][metrics].append(value)\n",
    "print(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(r'data\\Experiment 2\\experiment_2_results.json', 'w') as f:\n",
    "     json.dump(results, f, indent=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "t_test_result=perform_t_tests(results,label_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Two Nodes vs Three Nodes': {'Accuracy': {'t_stat': 3.3722302984300527, 'p_value': 0.0031991281850047024}, 'Sensitivity': {'t_stat': 4.419986126418913, 'p_value': 0.0002940485114338823}, 'Specificity': {'t_stat': 3.095569241656492, 'p_value': 0.005953620174863902}, 'AUC': {'t_stat': 7.610619014470917, 'p_value': 3.4881841597086767e-07}}, 'Two Nodes vs Four Nodes': {'Accuracy': {'t_stat': 10.91093078267511, 'p_value': 1.2694631178202622e-09}, 'Sensitivity': {'t_stat': 3.300322364695619, 'p_value': 0.003762617508787943}, 'Specificity': {'t_stat': 9.757157580260355, 'p_value': 7.808668639564761e-09}, 'AUC': {'t_stat': 13.68494391656638, 'p_value': 2.7328857386714388e-11}}, 'Two Nodes vs Five Nodes': {'Accuracy': {'t_stat': 15.835096397300699, 'p_value': 2.1143570527335817e-12}, 'Sensitivity': {'t_stat': 6.97878241880444, 'p_value': 1.1966886679677582e-06}, 'Specificity': {'t_stat': 15.330739192067655, 'p_value': 3.748010686223064e-12}, 'AUC': {'t_stat': 17.696153395010317, 'p_value': 2.9095882209879197e-13}}, 'Two Nodes vs Six Nodes': {'Accuracy': {'t_stat': 16.74661686808907, 'p_value': 7.812398923879322e-13}, 'Sensitivity': {'t_stat': 7.017530168319296, 'p_value': 1.1078322130053652e-06}, 'Specificity': {'t_stat': 16.27316422809355, 'p_value': 1.3023746998333996e-12}, 'AUC': {'t_stat': 23.608255520787363, 'p_value': 1.531332524337378e-15}}, 'Two Nodes vs Seven Nodes': {'Accuracy': {'t_stat': 21.346419372704965, 'p_value': 9.71190025790711e-15}, 'Sensitivity': {'t_stat': 8.928111066770635, 'p_value': 3.163332964883047e-08}, 'Specificity': {'t_stat': 20.563317442854665, 'p_value': 1.9209618260535713e-14}, 'AUC': {'t_stat': 26.25789131088477, 'p_value': 2.1490751804828407e-16}}, 'Two Nodes vs Eight Nodes': {'Accuracy': {'t_stat': 17.448368539433222, 'p_value': 3.747683496336537e-13}, 'Sensitivity': {'t_stat': 7.368809155584333, 'p_value': 5.555576782113498e-07}, 'Specificity': {'t_stat': 17.073185893797596, 'p_value': 5.532004750445883e-13}, 'AUC': {'t_stat': 27.981951737511125, 'p_value': 6.608660014830335e-17}}, 'Two Nodes vs Nine Nodes': {'Accuracy': {'t_stat': 24.54304670915985, 'p_value': 7.487473049376028e-16}, 'Sensitivity': {'t_stat': 8.612554537236784, 'p_value': 5.507807960209975e-08}, 'Specificity': {'t_stat': 23.910794844956644, 'p_value': 1.2113128544669878e-15}, 'AUC': {'t_stat': 27.09695790035408, 'p_value': 1.199857027548368e-16}}, 'Two Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 27.68063198759991, 'p_value': 8.080050129061877e-17}, 'Sensitivity': {'t_stat': 8.487117380789638, 'p_value': 6.89018342151183e-08}, 'Specificity': {'t_stat': 26.519892404837055, 'p_value': 1.788125168688295e-16}, 'AUC': {'t_stat': 30.631243777271557, 'p_value': 1.2277374605147683e-17}}, 'Three Nodes vs Four Nodes': {'Accuracy': {'t_stat': 7.0002202983709, 'p_value': 1.146652540494066e-06}, 'Sensitivity': {'t_stat': 1.131616426240755, 'p_value': 0.2718767200705809}, 'Specificity': {'t_stat': 6.653051434619935, 'p_value': 2.307462630543786e-06}, 'AUC': {'t_stat': 10.546452562805504, 'p_value': 2.218731386107909e-09}}, 'Three Nodes vs Five Nodes': {'Accuracy': {'t_stat': 14.459547211857716, 'p_value': 1.047696075537827e-11}, 'Sensitivity': {'t_stat': 3.568756885240619, 'p_value': 0.002048992008133004}, 'Specificity': {'t_stat': 14.051076783867217, 'p_value': 1.727523534394157e-11}, 'AUC': {'t_stat': 19.3718305863726, 'p_value': 5.684294543100556e-14}}, 'Three Nodes vs Six Nodes': {'Accuracy': {'t_stat': 16.72707959082278, 'p_value': 7.976893864817385e-13}, 'Sensitivity': {'t_stat': 4.758493237993915, 'p_value': 0.00013665646139587208}, 'Specificity': {'t_stat': 16.940157525592067, 'p_value': 6.362642272728022e-13}, 'AUC': {'t_stat': 24.415565561346096, 'p_value': 8.242317881292483e-16}}, 'Three Nodes vs Seven Nodes': {'Accuracy': {'t_stat': 26.137785751406955, 'p_value': 2.339434850168245e-16}, 'Sensitivity': {'t_stat': 5.494893413627583, 'p_value': 2.664536150148672e-05}, 'Specificity': {'t_stat': 25.12287832154757, 'p_value': 4.865689958975608e-16}, 'AUC': {'t_stat': 29.642351228278724, 'p_value': 2.2625901728887413e-17}}, 'Three Nodes vs Eight Nodes': {'Accuracy': {'t_stat': 17.363672008166468, 'p_value': 4.089380274481395e-13}, 'Sensitivity': {'t_stat': 4.830634645570104, 'p_value': 0.0001161821847681362}, 'Specificity': {'t_stat': 16.861340799932467, 'p_value': 6.915642048522918e-13}, 'AUC': {'t_stat': 27.013202181976734, 'p_value': 1.2707461267822694e-16}}, 'Three Nodes vs Nine Nodes': {'Accuracy': {'t_stat': 25.60243779110738, 'p_value': 3.430695268445449e-16}, 'Sensitivity': {'t_stat': 6.280969243314783, 'p_value': 4.97042877220684e-06}, 'Specificity': {'t_stat': 25.980709282794038, 'p_value': 2.6155289675321496e-16}, 'AUC': {'t_stat': 26.011701153131458, 'p_value': 2.558459028838005e-16}}, 'Three Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 28.03906123712165, 'p_value': 6.363095421352044e-17}, 'Sensitivity': {'t_stat': 6.720694521493637, 'p_value': 2.0109823455686853e-06}, 'Specificity': {'t_stat': 26.937342583502524, 'p_value': 1.3387560129893395e-16}, 'AUC': {'t_stat': 35.149319135681765, 'p_value': 9.388999089730174e-19}}, 'Four Nodes vs Five Nodes': {'Accuracy': {'t_stat': 7.299423903567193, 'p_value': 6.358591082201514e-07}, 'Sensitivity': {'t_stat': 2.1569647163135652, 'p_value': 0.04402712511019532}, 'Specificity': {'t_stat': 6.698123442105877, 'p_value': 2.1052714490752005e-06}, 'AUC': {'t_stat': 9.596616132743582, 'p_value': 1.0173355013336274e-08}}, 'Four Nodes vs Six Nodes': {'Accuracy': {'t_stat': 9.174781395435343, 'p_value': 2.068346479292692e-08}, 'Sensitivity': {'t_stat': 3.662690240434305, 'p_value': 0.0016546079768103778}, 'Specificity': {'t_stat': 8.639630990005074, 'p_value': 5.2492882676971695e-08}, 'AUC': {'t_stat': 15.984688789988777, 'p_value': 1.7895887606835222e-12}}, 'Four Nodes vs Seven Nodes': {'Accuracy': {'t_stat': 16.748213912968406, 'p_value': 7.799110839716231e-13}, 'Sensitivity': {'t_stat': 4.029033692425949, 'p_value': 0.0007170277426292308}, 'Specificity': {'t_stat': 15.676987014741488, 'p_value': 2.5256547759877673e-12}, 'AUC': {'t_stat': 21.600719897012972, 'p_value': 7.821093896828104e-15}}, 'Four Nodes vs Eight Nodes': {'Accuracy': {'t_stat': 12.166830012482068, 'p_value': 2.0512707458555996e-10}, 'Sensitivity': {'t_stat': 4.3339735134145325, 'p_value': 0.00035760801049412714}, 'Specificity': {'t_stat': 11.782238298550881, 'p_value': 3.5278172475472893e-10}, 'AUC': {'t_stat': 20.20068649688456, 'p_value': 2.6559233771745008e-14}}, 'Four Nodes vs Nine Nodes': {'Accuracy': {'t_stat': 18.196594962991156, 'p_value': 1.7617029178457167e-13}, 'Sensitivity': {'t_stat': 5.159238493540829, 'p_value': 5.577647205786201e-05}, 'Specificity': {'t_stat': 17.656196788549295, 'p_value': 3.0301595942397995e-13}, 'AUC': {'t_stat': 21.86796170064158, 'p_value': 6.245154307403206e-15}}, 'Four Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 20.055877383604287, 'p_value': 3.0272420219119293e-14}, 'Sensitivity': {'t_stat': 5.171472369774501, 'p_value': 5.428369062186481e-05}, 'Specificity': {'t_stat': 18.865188888940754, 'p_value': 9.184619938547492e-14}, 'AUC': {'t_stat': 29.95316619878747, 'p_value': 1.863167180573846e-17}}, 'Five Nodes vs Six Nodes': {'Accuracy': {'t_stat': 4.558838529513538, 'p_value': 0.00021456019338944568}, 'Sensitivity': {'t_stat': 1.8474303476044027, 'p_value': 0.08031635054574364}, 'Specificity': {'t_stat': 4.461614455294551, 'p_value': 0.00026750943893630503}, 'AUC': {'t_stat': 5.947981853441623, 'p_value': 1.0029244035496705e-05}}, 'Five Nodes vs Seven Nodes': {'Accuracy': {'t_stat': 8.66458435272877, 'p_value': 5.022210091415315e-08}, 'Sensitivity': {'t_stat': 2.184718706357749, 'p_value': 0.04164290640873728}, 'Specificity': {'t_stat': 8.576926163580364, 'p_value': 5.86831018896213e-08}, 'AUC': {'t_stat': 14.278624509796426, 'p_value': 1.3055437570234159e-11}}, 'Five Nodes vs Eight Nodes': {'Accuracy': {'t_stat': 9.066640527251398, 'p_value': 2.4895404789698622e-08}, 'Sensitivity': {'t_stat': 2.4522413731941106, 'p_value': 0.024035670971861746}, 'Specificity': {'t_stat': 9.05534810800657, 'p_value': 2.5384032318409606e-08}, 'AUC': {'t_stat': 13.207689537224285, 'p_value': 5.046363481560662e-11}}, 'Five Nodes vs Nine Nodes': {'Accuracy': {'t_stat': 12.317082939003745, 'p_value': 1.6657266860427636e-10}, 'Sensitivity': {'t_stat': 3.134099370647155, 'p_value': 0.005463238226309685}, 'Specificity': {'t_stat': 12.45702719130179, 'p_value': 1.374571592919537e-10}, 'AUC': {'t_stat': 14.562730064364738, 'p_value': 9.250982210112169e-12}}, 'Five Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 17.08904584217525, 'p_value': 5.440854568267969e-13}, 'Sensitivity': {'t_stat': 3.4617699250445013, 'p_value': 0.002612307981784326}, 'Specificity': {'t_stat': 17.389126148262413, 'p_value': 3.983379544528646e-13}, 'AUC': {'t_stat': 20.828896727582197, 'p_value': 1.5202482771257476e-14}}, 'Six Nodes vs Seven Nodes': {'Accuracy': {'t_stat': 5.166049979658655, 'p_value': 5.494023914855694e-05}, 'Sensitivity': {'t_stat': 0.6133765106080245, 'p_value': 0.5469009164840214}, 'Specificity': {'t_stat': 5.406545570164177, 'p_value': 3.232693133337584e-05}, 'AUC': {'t_stat': 7.28980617868401, 'p_value': 6.479035183228496e-07}}, 'Six Nodes vs Eight Nodes': {'Accuracy': {'t_stat': 5.814202154268624, 'p_value': 1.3349218749828532e-05}, 'Sensitivity': {'t_stat': 0.8002659808925354, 'p_value': 0.4334495075301513}, 'Specificity': {'t_stat': 5.960335747630659, 'p_value': 9.768962851785885e-06}, 'AUC': {'t_stat': 10.59091039088248, 'p_value': 2.0711087882219683e-09}}, 'Six Nodes vs Nine Nodes': {'Accuracy': {'t_stat': 8.400727563027912, 'p_value': 8.048553379178453e-08}, 'Sensitivity': {'t_stat': 2.0137306654157676, 'p_value': 0.058420807221130404}, 'Specificity': {'t_stat': 8.625663099751119, 'p_value': 5.381035683624471e-08}, 'AUC': {'t_stat': 13.243417988154837, 'p_value': 4.816905464355713e-11}}, 'Six Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 11.061456721683122, 'p_value': 1.012080907241226e-09}, 'Sensitivity': {'t_stat': 1.825039745092464, 'p_value': 0.08375864097674031}, 'Specificity': {'t_stat': 10.898826364432082, 'p_value': 1.2929354768722171e-09}, 'AUC': {'t_stat': 18.08305121283535, 'p_value': 1.9719533250739005e-13}}, 'Seven Nodes vs Eight Nodes': {'Accuracy': {'t_stat': 2.287559345290182, 'p_value': 0.033802649453682526}, 'Sensitivity': {'t_stat': 0.008324993507263971, 'p_value': 0.9934444835289794}, 'Specificity': {'t_stat': 2.252397826054129, 'p_value': 0.03631583757837285}, 'AUC': {'t_stat': 4.351130233633824, 'p_value': 0.00034390935875829325}}, 'Seven Nodes vs Nine Nodes': {'Accuracy': {'t_stat': 4.074389118135572, 'p_value': 0.0006464696252103176}, 'Sensitivity': {'t_stat': 1.593443685694234, 'p_value': 0.12756082266712118}, 'Specificity': {'t_stat': 4.193324892548918, 'p_value': 0.0004927637082866882}, 'AUC': {'t_stat': 6.226644518369568, 'p_value': 5.568150871444405e-06}}, 'Seven Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 7.027805312065034, 'p_value': 1.0854337689746547e-06}, 'Sensitivity': {'t_stat': 1.3388728126435478, 'p_value': 0.19641195271204973}, 'Specificity': {'t_stat': 6.848184437839415, 'p_value': 1.5544071439725074e-06}, 'AUC': {'t_stat': 12.614342715206437, 'p_value': 1.1098319066305293e-10}}, 'Eight Nodes vs Nine Nodes': {'Accuracy': {'t_stat': 0.7619047384852389, 'p_value': 0.4554712441220925}, 'Sensitivity': {'t_stat': 1.1942112446236075, 'p_value': 0.24709192948807457}, 'Specificity': {'t_stat': 0.8243032499385885, 'p_value': 0.4199954185482573}, 'AUC': {'t_stat': 4.14071757378648, 'p_value': 0.0005556198719773707}}, 'Eight Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 4.624056758116048, 'p_value': 0.00018510842903398625}, 'Sensitivity': {'t_stat': 1.3250381823911495, 'p_value': 0.20087376678506172}, 'Specificity': {'t_stat': 4.796976472705185, 'p_value': 0.00012531587121741996}, 'AUC': {'t_stat': 10.835792426379992, 'p_value': 1.4227139675332232e-09}}, 'Nine Nodes vs Ten Nodes': {'Accuracy': {'t_stat': 2.611839004296771, 'p_value': 0.017146251799811114}, 'Sensitivity': {'t_stat': 0.04071037977066938, 'p_value': 0.9679515384908166}, 'Specificity': {'t_stat': 2.7156740817419074, 'p_value': 0.013717151560364854}, 'AUC': {'t_stat': 5.174386060650387, 'p_value': 5.393420874851037e-05}}}\n"
     ]
    }
   ],
   "source": [
    "print(t_test_result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Two Nodes vs Three Nodes\n",
      "{'t_stat': 3.3722302984300527, 'p_value': 0.0031991281850047024}\n",
      "Two Nodes vs Four Nodes\n",
      "{'t_stat': 10.91093078267511, 'p_value': 1.2694631178202622e-09}\n",
      "Two Nodes vs Five Nodes\n",
      "{'t_stat': 15.835096397300699, 'p_value': 2.1143570527335817e-12}\n",
      "Two Nodes vs Six Nodes\n",
      "{'t_stat': 16.74661686808907, 'p_value': 7.812398923879322e-13}\n",
      "Two Nodes vs Seven Nodes\n",
      "{'t_stat': 21.346419372704965, 'p_value': 9.71190025790711e-15}\n",
      "Two Nodes vs Eight Nodes\n",
      "{'t_stat': 17.448368539433222, 'p_value': 3.747683496336537e-13}\n",
      "Two Nodes vs Nine Nodes\n",
      "{'t_stat': 24.54304670915985, 'p_value': 7.487473049376028e-16}\n",
      "Two Nodes vs Ten Nodes\n",
      "{'t_stat': 27.68063198759991, 'p_value': 8.080050129061877e-17}\n",
      "Three Nodes vs Four Nodes\n",
      "{'t_stat': 7.0002202983709, 'p_value': 1.146652540494066e-06}\n",
      "Three Nodes vs Five Nodes\n",
      "{'t_stat': 14.459547211857716, 'p_value': 1.047696075537827e-11}\n",
      "Three Nodes vs Six Nodes\n",
      "{'t_stat': 16.72707959082278, 'p_value': 7.976893864817385e-13}\n",
      "Three Nodes vs Seven Nodes\n",
      "{'t_stat': 26.137785751406955, 'p_value': 2.339434850168245e-16}\n",
      "Three Nodes vs Eight Nodes\n",
      "{'t_stat': 17.363672008166468, 'p_value': 4.089380274481395e-13}\n",
      "Three Nodes vs Nine Nodes\n",
      "{'t_stat': 25.60243779110738, 'p_value': 3.430695268445449e-16}\n",
      "Three Nodes vs Ten Nodes\n",
      "{'t_stat': 28.03906123712165, 'p_value': 6.363095421352044e-17}\n",
      "Four Nodes vs Five Nodes\n",
      "{'t_stat': 7.299423903567193, 'p_value': 6.358591082201514e-07}\n",
      "Four Nodes vs Six Nodes\n",
      "{'t_stat': 9.174781395435343, 'p_value': 2.068346479292692e-08}\n",
      "Four Nodes vs Seven Nodes\n",
      "{'t_stat': 16.748213912968406, 'p_value': 7.799110839716231e-13}\n",
      "Four Nodes vs Eight Nodes\n",
      "{'t_stat': 12.166830012482068, 'p_value': 2.0512707458555996e-10}\n",
      "Four Nodes vs Nine Nodes\n",
      "{'t_stat': 18.196594962991156, 'p_value': 1.7617029178457167e-13}\n",
      "Four Nodes vs Ten Nodes\n",
      "{'t_stat': 20.055877383604287, 'p_value': 3.0272420219119293e-14}\n",
      "Five Nodes vs Six Nodes\n",
      "{'t_stat': 4.558838529513538, 'p_value': 0.00021456019338944568}\n",
      "Five Nodes vs Seven Nodes\n",
      "{'t_stat': 8.66458435272877, 'p_value': 5.022210091415315e-08}\n",
      "Five Nodes vs Eight Nodes\n",
      "{'t_stat': 9.066640527251398, 'p_value': 2.4895404789698622e-08}\n",
      "Five Nodes vs Nine Nodes\n",
      "{'t_stat': 12.317082939003745, 'p_value': 1.6657266860427636e-10}\n",
      "Five Nodes vs Ten Nodes\n",
      "{'t_stat': 17.08904584217525, 'p_value': 5.440854568267969e-13}\n",
      "Six Nodes vs Seven Nodes\n",
      "{'t_stat': 5.166049979658655, 'p_value': 5.494023914855694e-05}\n",
      "Six Nodes vs Eight Nodes\n",
      "{'t_stat': 5.814202154268624, 'p_value': 1.3349218749828532e-05}\n",
      "Six Nodes vs Nine Nodes\n",
      "{'t_stat': 8.400727563027912, 'p_value': 8.048553379178453e-08}\n",
      "Six Nodes vs Ten Nodes\n",
      "{'t_stat': 11.061456721683122, 'p_value': 1.012080907241226e-09}\n",
      "Seven Nodes vs Eight Nodes\n",
      "{'t_stat': 2.287559345290182, 'p_value': 0.033802649453682526}\n",
      "Seven Nodes vs Nine Nodes\n",
      "{'t_stat': 4.074389118135572, 'p_value': 0.0006464696252103176}\n",
      "Seven Nodes vs Ten Nodes\n",
      "{'t_stat': 7.027805312065034, 'p_value': 1.0854337689746547e-06}\n",
      "Eight Nodes vs Nine Nodes\n",
      "{'t_stat': 0.7619047384852389, 'p_value': 0.4554712441220925}\n",
      "Eight Nodes vs Ten Nodes\n",
      "{'t_stat': 4.624056758116048, 'p_value': 0.00018510842903398625}\n",
      "Nine Nodes vs Ten Nodes\n",
      "{'t_stat': 2.611839004296771, 'p_value': 0.017146251799811114}\n",
      "Data successfully exported\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import openpyxl\n",
    "\n",
    "# Create a list of rows, where each row is a dict that will become one row in the final DataFrame\n",
    "rows = []\n",
    "\n",
    "for shared_label, metrics_dict in t_test_result.items():\n",
    "    print(shared_label)\n",
    "    print(metrics_dict['Accuracy'])\n",
    "    # Each \"metrics_dict\" has keys like 'Accuracy', 'Sensitivity', 'Specificity', 'AUC'.\n",
    "    # Each of these keys corresponds to a list of 20 floats (in your example).\n",
    "    for metric_name, metric_values in metrics_dict.items():\n",
    "        # metric_values should be {\"t_stat\": X, \"p_value\": Y}\n",
    "        row = {\n",
    "            \"Shared Label\": shared_label,\n",
    "            \"Metric\": metric_name,\n",
    "            \"t_stat\": metric_values[\"t_stat\"],\n",
    "            \"p_value\": metric_values[\"p_value\"]\n",
    "        }\n",
    "        rows.append(row)\n",
    "# Convert the list of dicts into a pandas DataFrame\n",
    "df = pd.DataFrame(rows)\n",
    "\n",
    "# Now export this DataFrame to Excel\n",
    "df.to_excel(r\"data\\t-tests\\Experiment 2 Result.xlsx\", index=False)\n",
    "\n",
    "print(\"Data successfully exported\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Experiment 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Two Shared': {'Accuracy': [0.9140518417462484, 0.9151012698079545, 0.9036625039353553, 0.9039773323538671, 0.8939028229614862, 0.8965263931157519, 0.8935879945429741, 0.8900199391331723, 0.9139468989400776, 0.8964214503095813, 0.9051317032217442, 0.9132122992968832, 0.9070206737328157, 0.9145765557771013, 0.8929583377059503, 0.892433623675097, 0.8844579704061287, 0.9049218176094029, 0.923706579913947, 0.9045020463847204], 'Sensitivity': [0.9506422892645344, 0.9079662580051664, 0.9518456824214727, 0.9405480045762906, 0.96327278632014, 0.9500368954774238, 0.9515469688329875, 0.9394697005893081, 0.9210879834331879, 0.9199341390138087, 0.9117455324785195, 0.9576787835955188, 0.9225864350716274, 0.938359581874106, 0.9189015644345049, 0.9416401142273275, 0.9246389866020812, 0.9342984238012302, 0.9216056604732876, 0.9307506936942371], 'Specificity': [0.914078213386522, 0.9156993381604099, 0.9018207324748558, 0.9033001810054204, 0.8930607585056325, 0.8940470560191265, 0.8929212133967523, 0.8907019224195385, 0.9133558491225873, 0.8930053540197702, 0.9080830880608934, 0.9128435651629742, 0.9086320881132578, 0.9146240132936743, 0.8917448976214223, 0.8886705984761725, 0.8837954500551253, 0.9038615254523257, 0.9253267674391771, 0.9034968427500181], 'AUC': [0.9654542092894242, 0.9525907188898167, 0.9667010175404147, 0.959078526589841, 0.9639104927807978, 0.9614123917310194, 0.9585064518365329, 0.9527147599095932, 0.9568510485428882, 0.9473138898178868, 0.951856379309674, 0.9668698473039056, 0.9555566147797057, 0.9634715678831564, 0.9407900837504257, 0.9528119208233365, 0.9493347006639428, 0.9556852938816783, 0.9605240539135499, 0.959116527812104]}, 'Three Shared': {'Accuracy': [0.9000944485255535, 0.9152062126141255, 0.9071256165389863, 0.8906495959701962, 0.9007241053625775, 0.9094343582747404, 0.9052366460279148, 0.8910693671948787, 0.8986252492391646, 0.8855073984678351, 0.9018784762304544, 0.9218176094028754, 0.9097491866932521, 0.9125826424598594, 0.8604260677930528, 0.9041872179662085, 0.8820442858642039, 0.8981005352083115, 0.9144716129709309, 0.9284290061916256], 'Sensitivity': [0.9549870616351371, 0.924936150722201, 0.9374067614893835, 0.9164083784181714, 0.9462472039044921, 0.9489709222549867, 0.9320207927261842, 0.9354534585542414, 0.9511954583277086, 0.937680095561843, 0.9321427863269445, 0.935001734652847, 0.9351778931968846, 0.9471481215082022, 0.9477327602623071, 0.9279567128955126, 0.9255181931598095, 0.9393728312461461, 0.9250536941154862, 0.9362223074493182], 'Specificity': [0.8982837283755672, 0.9155688880895225, 0.90901468820578, 0.8900146964549867, 0.9004845534488987, 0.9094325842146008, 0.9045828516340237, 0.890242684765092, 0.8969441229586533, 0.8833425826696997, 0.9054531309927837, 0.9227341693068976, 0.9093118198269341, 0.9095683010641713, 0.8576834760343934, 0.9011856942784102, 0.8799611680834223, 0.8949783701207414, 0.9124395958924383, 0.9276840645071621], 'AUC': [0.9640559732635418, 0.9573908840871787, 0.9634575409908357, 0.9475150008182537, 0.9603318189768408, 0.9641285159346326, 0.9537573903607159, 0.9507249187789284, 0.9567678139572743, 0.9538314985819564, 0.9588877301236214, 0.9647887748940812, 0.9556189105308159, 0.9649116979491773, 0.9453113677446865, 0.9553951033777358, 0.952862967678215, 0.9576244505861767, 0.9603291203646661, 0.9634012806906214]}, 'Four Shared': {'Accuracy': [0.9186693252177563, 0.9176198971560499, 0.8962115646972401, 0.9088047014377163, 0.9180396683807325, 0.9019834190366249, 0.9110085003672999, 0.8710252912162871, 0.8929583377059502, 0.9040822751600379, 0.8795256585161086, 0.9240214083324589, 0.9096442438870814, 0.9262252072620422, 0.8820442858642041, 0.8892853394899779, 0.9157309266449786, 0.8865568265295413, 0.8976807639836287, 0.9298982054780145], 'Sensitivity': [0.9310295181977383, 0.9285959009316075, 0.9426952590001031, 0.9446350915596397, 0.9254997658656603, 0.9590826062219686, 0.9384797793968314, 0.9446978715778506, 0.945547405671149, 0.9284836153278339, 0.9281761890218585, 0.9281495921253087, 0.9339448656614764, 0.9253470821273526, 0.9408417660127154, 0.9646382236106645, 0.9188467009925881, 0.9505717893257866, 0.9386102485148637, 0.9370782520826553], 'Specificity': [0.917875119466795, 0.9157585499689824, 0.8940188467522245, 0.9074367856639248, 0.9189429252962188, 0.8991341997623105, 0.9115488909180165, 0.8699119727377838, 0.8901360707604299, 0.9032852403595892, 0.879709606143013, 0.9276943404407952, 0.9099729135352644, 0.9275748949887467, 0.8783147625295586, 0.8844884644179274, 0.9166742459827868, 0.8856649947081597, 0.8992187213624361, 0.9301910782843239], 'AUC': [0.9616062106265327, 0.9575056475663837, 0.9637023454483534, 0.9620865504376733, 0.9589650594196052, 0.9627339291774295, 0.9584366632439253, 0.9337777722962577, 0.9545532389347227, 0.9519773567615372, 0.9488337891823189, 0.9650134229278784, 0.9541470621000462, 0.9650951487864101, 0.9524241649011833, 0.960661754175373, 0.9549531928776205, 0.9603147264707376, 0.9588117189511929, 0.9672040249479857]}, 'Five Shared': {'Accuracy': [0.8989400776576766, 0.9051317032217442, 0.9090145870500579, 0.8857172840801761, 0.9188792108300977, 0.9037674467415259, 0.9093294154685696, 0.8857172840801761, 0.8754328890754538, 0.893063280512121, 0.9039773323538671, 0.935565117011229, 0.9128974708783714, 0.9089096442438873, 0.8987301920453353, 0.8991499632700178, 0.8838283135691049, 0.9066009025081331, 0.9266449784867248, 0.9048168748032323], 'Sensitivity': [0.954983144409949, 0.9171894074922734, 0.9306252849094747, 0.9550970003604886, 0.9309112841645046, 0.9553296345747169, 0.9355040763199111, 0.9393658651201453, 0.9456090677321534, 0.9466363986497505, 0.921685031715964, 0.935889165637233, 0.8906890203898227, 0.9526035236417751, 0.9428625953618356, 0.9515175940959494, 0.9279352389324342, 0.9310980501201187, 0.9254714187300459, 0.9598224727894158], 'Specificity': [0.8980704561508351, 0.9044902464987654, 0.9051072583239024, 0.8792790425638564, 0.9198751549508912, 0.9018155239003172, 0.9097012750422699, 0.8841075616739984, 0.872848169097125, 0.8899891929194089, 0.9051794626025707, 0.9377970293573588, 0.9151518792776041, 0.9079081769386097, 0.8958244076605214, 0.8978288013307162, 0.8806545550640454, 0.9062964360379155, 0.9297668864776656, 0.9009984672438331], 'AUC': [0.965364151750459, 0.9480945033827115, 0.9597205568378053, 0.9592187166553852, 0.9603645582323643, 0.9603484895108535, 0.9587476695274065, 0.9466287580625715, 0.9493473711361708, 0.9551727285340946, 0.9550407890748421, 0.9698010267099512, 0.9445106262791315, 0.9669805675742393, 0.9508469847826386, 0.9616965400389919, 0.9519194298116388, 0.9575654352990952, 0.9656850069970186, 0.9628341367329339]}, 'Six Shared': {'Accuracy': [0.9205582957288279, 0.9223423234337287, 0.9133172421030538, 0.92097806695351, 0.9143666701647603, 0.9034526183230139, 0.8903347675516844, 0.8757477174939658, 0.8977857067897994, 0.8961066218910693, 0.9130024136845419, 0.9242312939448001, 0.9080701017945221, 0.907230559345157, 0.9151012698079548, 0.9100640151117643, 0.8943225941861684, 0.9061811312834505, 0.8969461643404344, 0.8975758211774584], 'Sensitivity': [0.9418366801590943, 0.9201837356759067, 0.9504644824114311, 0.9116724510510248, 0.9501916320658469, 0.9512097343153738, 0.9570950167508392, 0.953041952333042, 0.9229499034587788, 0.9275065775130604, 0.9107721331180235, 0.9421172312840126, 0.9277788691940856, 0.9385591224820448, 0.889986698284294, 0.9403432689458219, 0.9190224321201137, 0.9506557022639363, 0.9560678397650473, 0.9507052556272751], 'Specificity': [0.9196496639375312, 0.9224104185440818, 0.9116804092859996, 0.9249230693997337, 0.9158676607329895, 0.9027950742726234, 0.8866656281992515, 0.8735699442740883, 0.8967726104360951, 0.8940724594153274, 0.9168426898817065, 0.9260972338428038, 0.9099072318761977, 0.9068634675160733, 0.9167302708615215, 0.9093106916965812, 0.8948454750925018, 0.9030743854328411, 0.8958241650896388, 0.8950187818291693], 'AUC': [0.9655167567062045, 0.9578678513422815, 0.9677404462851329, 0.959113144569289, 0.9659804782914129, 0.9600010437985586, 0.9552667886733277, 0.9411914682709581, 0.9523992856561114, 0.9521857211709969, 0.9527107973994599, 0.9679416996394823, 0.9521818355153605, 0.9601033740923094, 0.9443250323737852, 0.9625731466163688, 0.9524945767021982, 0.959177301574275, 0.961805831642126, 0.9599048931159538]}, 'Seven Shared': {'Accuracy': [0.9104837863364467, 0.9229719802707526, 0.902613075873649, 0.9119529856228356, 0.915311155420296, 0.9089096442438872, 0.8928533948997794, 0.9119529856228356, 0.8946374226046804, 0.9093294154685696, 0.8979955924021409, 0.9033476755168433, 0.91268758526603, 0.9179347255745618, 0.8899149963270018, 0.8875013117850773, 0.88141462902718, 0.9168852975128554, 0.8971560499527758, 0.903137789904502], 'Sensitivity': [0.9438487047450074, 0.928199506030976, 0.9483542377604375, 0.9237593835428259, 0.9362918968872377, 0.9523216895316888, 0.9512973103424714, 0.9398686063110613, 0.9202939568613211, 0.9028985908890407, 0.9235109841550406, 0.9465098346080608, 0.9391753474266066, 0.9316402582148916, 0.946603135867082, 0.9565486550498025, 0.936371776439199, 0.9191633905198208, 0.9321935206562895, 0.9627871748560491], 'Specificity': [0.9083086285119408, 0.9219289954089193, 0.9006086111274326, 0.913687162019985, 0.9147974771800376, 0.9082521856905343, 0.8914587064480275, 0.9109393458906276, 0.8895478438900722, 0.9091199293644829, 0.8981372569883928, 0.9026069239148792, 0.9132856722436723, 0.9183389022088259, 0.8880199765378478, 0.8840998329985427, 0.8826934850653617, 0.9157324109830811, 0.896376941759509, 0.8991607771261342], 'AUC': [0.9638656711631227, 0.9599444420334818, 0.965026339993672, 0.9594202471390063, 0.9575698103561578, 0.9650534445202582, 0.9608014428085462, 0.9568552064109266, 0.9493648406895983, 0.944593575400267, 0.952540210203554, 0.961722384037755, 0.9571873516188186, 0.9628014298502336, 0.9531638483912649, 0.9543857728369746, 0.9515055900800908, 0.9533728066814839, 0.9605469091884358, 0.9658616648488386]}, 'Eight Shared': {'Accuracy': [0.9019834190366249, 0.9178297827683912, 0.8950571938293629, 0.8969461643404344, 0.9136320705215658, 0.8993598488823591, 0.904292160772379, 0.8861370553048589, 0.8866617693357118, 0.8954769650540455, 0.9097491866932523, 0.9275894637422605, 0.9223423234337287, 0.9047119319970616, 0.8928533948997797, 0.8687165494805332, 0.8826739427012279, 0.9069157309266449, 0.9060761884772799, 0.9243362367509709], 'Sensitivity': [0.9537060721754369, 0.9251486670641745, 0.9551474948722212, 0.9365649732553731, 0.9435420889353329, 0.9534374082229065, 0.9491193781759445, 0.9333245843616305, 0.9443986246139547, 0.9303501614783147, 0.9334227352041096, 0.9328511910547618, 0.9321494620554938, 0.9484063875061862, 0.935107718809684, 0.9640112647321027, 0.9378365067625217, 0.9453282710077718, 0.9494834900488296, 0.9505203834136451], 'Specificity': [0.8969285908585108, 0.9157791534696298, 0.8915913724702051, 0.8984089674623217, 0.9151790179346482, 0.8978445737744837, 0.9019899887609143, 0.8845475051103553, 0.8843799664358735, 0.8942345706116119, 0.9099282057032607, 0.9294596995862652, 0.9235229008380028, 0.9058033857536432, 0.8921778736783968, 0.8655355428125068, 0.8792718590846894, 0.9039830864199712, 0.9056974414054918, 0.9236998815604278], 'AUC': [0.9646836470783676, 0.9562136880500203, 0.9643457332957002, 0.9586735202323637, 0.9649502847003405, 0.9613686370753064, 0.9587683172033302, 0.9494204700596192, 0.9519710720564325, 0.9519692414216261, 0.9581541888238245, 0.9665629578439987, 0.9603199680066496, 0.9654875273193384, 0.9524552009624572, 0.9539185514686734, 0.955519478108141, 0.9578944105531648, 0.9630539889105125, 0.9694953198162515]}, 'Nine Shared': {'Accuracy': [0.9167803547066848, 0.9052366460279148, 0.9030328470983314, 0.8997796201070417, 0.9288487774163079, 0.9196138104732922, 0.9131073564907126, 0.8785811732605728, 0.9094343582747402, 0.893902822961486, 0.9085948158253753, 0.9269598069052367, 0.9132122992968833, 0.8984153636268234, 0.8812047434148391, 0.9030328470983313, 0.8906495959701962, 0.9146814985832722, 0.9202434673103159, 0.9338860321124987], 'Sensitivity': [0.9435754500754192, 0.9452991718971174, 0.9571988137048251, 0.9415172109270719, 0.9270535984982066, 0.9420122865221431, 0.941903736700753, 0.9443077537294933, 0.9317313498683848, 0.9393465654661608, 0.9264574770126028, 0.9373776845912545, 0.9210054387833811, 0.9613243493024914, 0.9349513683059473, 0.9560805107249416, 0.9414296151750127, 0.9348587286476003, 0.9286605598420492, 0.9222072429539317], 'Specificity': [0.9173395422117179, 0.9011152999604751, 0.8996701961605506, 0.8986030305183176, 0.9295844540843177, 0.9197681716457888, 0.9134612910234872, 0.8769839880261695, 0.9100493616262673, 0.8932460632339474, 0.9090112739205926, 0.9278760109564784, 0.9143069742344697, 0.8970385699662542, 0.8802973596249465, 0.9033298071792311, 0.8906132304992862, 0.9128919307152239, 0.9237398763276347, 0.9341810405521394], 'AUC': [0.9670341557131075, 0.9555277778853235, 0.9703891762380573, 0.9598508135283821, 0.9633365250997982, 0.9621079476267137, 0.961052306337499, 0.9459630225032766, 0.9570916365277269, 0.9530291133452837, 0.9592148656448294, 0.9652290011773836, 0.9540497418127811, 0.964280798189357, 0.9484502486935544, 0.9656870097713212, 0.9579273126232022, 0.9617177299112806, 0.9657676035770849, 0.9641453064228404]}, 'Ten Shared': {'Accuracy': [0.9112183859796411, 0.9104837863364467, 0.903452618323014, 0.9018784762304546, 0.9092244726623989, 0.8845629132122994, 0.9131073564907125, 0.897680763983629, 0.9067058453143038, 0.8928533948997797, 0.9007241053625774, 0.9262252072620422, 0.9098541294994228, 0.8932731661244622, 0.8944275369923391, 0.8911743100010495, 0.9071256165389863, 0.9168852975128556, 0.9153111554202957, 0.9140518417462482], 'Sensitivity': [0.9522984377069896, 0.9217614364672758, 0.9418049777989084, 0.9386321412088633, 0.9523598350905872, 0.9722701917055693, 0.9421419574889831, 0.924197906707486, 0.937866040815795, 0.9389750559989783, 0.9229253466742419, 0.947445749176343, 0.9493090013953484, 0.952346532947626, 0.946392480415662, 0.9705060134667411, 0.9192509633185602, 0.9419755062705713, 0.9267109814169443, 0.9603928236443587], 'Specificity': [0.9123802127096504, 0.9092130065266942, 0.9015685043790723, 0.9029774357284851, 0.9074103294735079, 0.8802133777585144, 0.913558657410459, 0.8966013625017248, 0.9062521587585307, 0.8907218034131905, 0.9046677227121914, 0.926991140856351, 0.90975159374351, 0.8924812928460255, 0.8924643771333345, 0.8858273496031875, 0.906869500324131, 0.915138861381975, 0.9166367833224954, 0.9114185973803036], 'AUC': [0.9672219431480579, 0.9586501232097556, 0.9635640905985077, 0.9597279660089354, 0.9647399021104525, 0.9612085389811468, 0.9611760436577038, 0.9462757248652223, 0.9588953299607262, 0.9524886437504239, 0.9555023886144768, 0.9691832839998666, 0.9625186588660519, 0.9606868303415439, 0.9538789607099477, 0.9621890662049009, 0.9567029738054216, 0.9616692919311493, 0.9637222354085856, 0.9661937621514436]}, 'Eleven Shared': {'Accuracy': [0.9277993493546017, 0.9110085003672999, 0.9010389337810893, 0.9070206737328156, 0.9211879525658516, 0.9085948158253753, 0.9063910168957917, 0.8864518837233708, 0.8923286808689265, 0.8972609927589463, 0.895896736278728, 0.921397838178193, 0.9225522090460697, 0.8985203064329941, 0.8894952251023193, 0.8991499632700177, 0.899674677300871, 0.8975758211774584, 0.9315772903767445, 0.9137370133277364], 'Sensitivity': [0.9288880968477489, 0.9386972555553859, 0.952071073253507, 0.9368384140131608, 0.93681565090405, 0.958171913663765, 0.9489927102151879, 0.9641515956616437, 0.9317877277804733, 0.9252269345010362, 0.927047948982328, 0.9545243146354342, 0.9406135440293096, 0.961641490330015, 0.9402308049945147, 0.965644089303012, 0.93338170860635, 0.9546673477600822, 0.9230735950735391, 0.9505180891240003], 'Specificity': [0.9294793110012081, 0.9099130481596542, 0.8998631141937392, 0.9073311900743631, 0.9213552019935013, 0.9068133274578762, 0.9038104185172635, 0.883841383633135, 0.8920433989660174, 0.8991694709751558, 0.8969591163853452, 0.92120424918396, 0.9222782448936998, 0.8958510312479504, 0.888341131260694, 0.8958082473974347, 0.9010917774473953, 0.8937763206234586, 0.9330285826282249, 0.9132086962579221], 'AUC': [0.9642288282042315, 0.9581671228332126, 0.9664285032563215, 0.9609438909328294, 0.962799626433769, 0.9666946014867392, 0.9600752685980626, 0.9582589350175549, 0.9558773569833215, 0.9540681433181016, 0.9565190634226629, 0.9697526028584386, 0.9637173510193882, 0.965793888183208, 0.9509898634059601, 0.9654365918041924, 0.9583355477501819, 0.9608112301803515, 0.967525453242162, 0.9649214294654596]}, 'Twelve Shared': {'Accuracy': [0.9131073564907125, 0.9061811312834506, 0.9082799874068631, 0.9106936719487881, 0.9180396683807325, 0.9090145870500579, 0.9072305593451568, 0.8815195718333508, 0.8908594815825374, 0.8758526603001364, 0.9198236960856333, 0.9411270857382726, 0.9217126665967049, 0.9110085003672999, 0.9141567845524189, 0.9049218176094028, 0.9006191625564066, 0.9109035575611292, 0.9159408122573198, 0.9127925280722007], 'Sensitivity': [0.9448685795861711, 0.9214529337165557, 0.9331975896755316, 0.9340303037477392, 0.9552692188469929, 0.941266755808381, 0.9538392426904385, 0.9564711645043941, 0.9501391089248086, 0.954926337631375, 0.9227953691249093, 0.9382503718237025, 0.9255923920243722, 0.962579078986902, 0.9150390373160853, 0.9614753222071181, 0.9253610234163958, 0.9404776041275239, 0.94309252536779, 0.9401770350497112], 'Specificity': [0.9115524018693114, 0.9032220189097668, 0.9077023303790299, 0.9125162229724697, 0.917286224286532, 0.907683605928302, 0.9045715242782152, 0.8804241333923528, 0.8882410531814227, 0.8725572468800261, 0.921019066424567, 0.94390568738854, 0.9245530551748368, 0.9082738087536509, 0.9150806502873622, 0.9006780732886727, 0.9021810963775531, 0.9108004587291745, 0.9151341700840749, 0.9122017449419375], 'AUC': [0.9669643702260001, 0.9534165081767438, 0.9631268813010586, 0.9610534025859258, 0.9686848757803102, 0.9599227587847261, 0.9600687217352963, 0.9502502328823925, 0.9575702655992627, 0.9520017875857338, 0.9579347216303282, 0.972694562109345, 0.958975609633199, 0.9689911669189464, 0.9525052446702398, 0.9645747670398795, 0.9555329187949844, 0.9617092749417574, 0.9659527818212609, 0.9618924231151134]}, 'Thirteen Shared': {'Accuracy': [0.9021933046489663, 0.9107986147549585, 0.9121628712351768, 0.8990450204638474, 0.929478434253332, 0.9019834190366249, 0.9013537621996011, 0.888970511071466, 0.8972609927589463, 0.8926435092874385, 0.8885507398467835, 0.9293734914471611, 0.9116381572043235, 0.9227620946584113, 0.888970511071466, 0.8954769650540455, 0.9003043341378947, 0.9196138104732919, 0.911533214398153, 0.9316822331829152], 'Sensitivity': [0.9593016956118187, 0.9362995272065706, 0.9496340522585663, 0.9378140054296158, 0.9305775512249974, 0.9629802766821793, 0.9540100459888271, 0.9449327685584988, 0.9254751711129092, 0.9417863027138539, 0.9340866763839586, 0.9447035464135004, 0.9308708917505955, 0.9348317412021755, 0.9503461061630338, 0.9715581825174497, 0.9253538196491833, 0.9395496593907807, 0.9468463631021304, 0.9410136285743204], 'Specificity': [0.9006425526902395, 0.9067712307974803, 0.9107536533719843, 0.8995007549530416, 0.9306383194650204, 0.9000150817927673, 0.8987033331309087, 0.8876788684646677, 0.8965604207133531, 0.8894217941322975, 0.890215058342317, 0.9305434231718118, 0.9120844259782989, 0.9236617206907307, 0.8864864747055149, 0.8890312571558365, 0.9029115954772552, 0.9186387541703808, 0.9116553180707585, 0.9327544185387595], 'AUC': [0.9654387189673902, 0.9585647230108892, 0.9670052545882963, 0.9616799667875432, 0.9641264288139203, 0.9657904577145509, 0.9600637010240356, 0.9532129552414449, 0.9529918694542464, 0.953901078695721, 0.9552284301846001, 0.9702605175309, 0.9555341382763097, 0.9648719894824567, 0.9545833389875379, 0.967746968144453, 0.9560002305187021, 0.9605268864025754, 0.9654909319019944, 0.9691139383820517]}}\n"
     ]
    }
   ],
   "source": [
    "import glob\n",
    "import os\n",
    "import json\n",
    "experiment_dir='data\\Experiment 3'\n",
    "results={}\n",
    "metrics=['Accuracy', 'Sensitivity', 'Specificity', 'AUC']\n",
    "label_list= ['Two Shared', 'Three Shared', 'Four Shared', 'Five Shared', 'Six Shared', 'Seven Shared', 'Eight Shared', 'Nine Shared', 'Ten Shared','Eleven Shared', 'Twelve Shared', 'Thirteen Shared']\n",
    "results = {\n",
    "    label: {\n",
    "        \"Accuracy\": [],\n",
    "        \"Sensitivity\": [],\n",
    "        \"Specificity\": [],\n",
    "        \"AUC\": []\n",
    "    } for label in label_list\n",
    "}\n",
    "for count in range(1,21):\n",
    "    with open(experiment_dir+'\\iteration_'+str(count)+'\\iteration_summary.json', 'r') as f:\n",
    "        data = json.load(f)\n",
    "    for item,label_name in zip(data,label_list):\n",
    "        for metrics, value in item['metrics']['Overall'].items():\n",
    "            results[label_name][metrics].append(value)\n",
    "print(results)\n",
    "\n",
    "\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(r'data\\Experiment 3\\experiment_3_results.json', 'w') as f:\n",
    "     json.dump(results, f, indent=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "t_test_result=perform_t_tests(results,label_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Two Shared vs Three Shared': {'Accuracy': {'t_stat': 0.4718886775693806, 'p_value': 0.6423803718375879}, 'Sensitivity': {'t_stat': -0.5152427694801677, 'p_value': 0.6123262479854461}, 'Specificity': {'t_stat': 0.5625654964687882, 'p_value': 0.5803085830331236}, 'AUC': {'t_stat': -0.5338801325324352, 'p_value': 0.5996157281965482}}, 'Two Shared vs Four Shared': {'Accuracy': {'t_stat': -0.22105680368423875, 'p_value': 0.827406451989789}, 'Sensitivity': {'t_stat': -0.6976407847283955, 'p_value': 0.493851476981242}, 'Specificity': {'t_stat': -0.1764411418662618, 'p_value': 0.861814487338039}, 'AUC': {'t_stat': -0.4230811199268106, 'p_value': 0.676986226883441}}, 'Two Shared vs Five Shared': {'Accuracy': {'t_stat': -0.031589774029780936, 'p_value': 0.9751286830114254}, 'Sensitivity': {'t_stat': -0.6259999929261341, 'p_value': 0.5387631587874306}, 'Specificity': {'t_stat': 0.14949236195972304, 'p_value': 0.8827405579721557}, 'AUC': {'t_stat': -0.36631128989588413, 'p_value': 0.7181800727545524}}, 'Two Shared vs Six Shared': {'Accuracy': {'t_stat': -1.1097725596972379, 'p_value': 0.2809459296218739}, 'Sensitivity': {'t_stat': -0.1977893927555281, 'p_value': 0.8453105826962402}, 'Specificity': {'t_stat': -1.072226899861069, 'p_value': 0.29704897573336525}, 'AUC': {'t_stat': -0.5005512237471748, 'p_value': 0.6224356687167569}}, 'Two Shared vs Seven Shared': {'Accuracy': {'t_stat': -0.4560457772804352, 'p_value': 0.6535266919293146}, 'Sensitivity': {'t_stat': -0.608664960463308, 'p_value': 0.5499550038302061}, 'Specificity': {'t_stat': -0.23604323791645257, 'p_value': 0.8159249348024875}, 'AUC': {'t_stat': -0.707600014302637, 'p_value': 0.48778338827760337}}, 'Two Shared vs Eight Shared': {'Accuracy': {'t_stat': 0.34932346211127724, 'p_value': 0.7306882580734704}, 'Sensitivity': {'t_stat': -2.47926912133182, 'p_value': 0.022710251459676922}, 'Specificity': {'t_stat': 0.526145579263656, 'p_value': 0.6048749644448206}, 'AUC': {'t_stat': -2.367853304635316, 'p_value': 0.028654175785618796}}, 'Two Shared vs Nine Shared': {'Accuracy': {'t_stat': -1.5186655749608875, 'p_value': 0.1453116962094483}, 'Sensitivity': {'t_stat': -1.0811952392409714, 'p_value': 0.29314307871065143}, 'Specificity': {'t_stat': -1.4779326815779261, 'p_value': 0.15581195560321207}, 'AUC': {'t_stat': -3.1051857025618967, 'p_value': 0.005827363832730752}}, 'Two Shared vs Ten Shared': {'Accuracy': {'t_stat': -0.7298273915688733, 'p_value': 0.474398389244731}, 'Sensitivity': {'t_stat': -2.3927051078820067, 'p_value': 0.027214650284337113}, 'Specificity': {'t_stat': -0.5954556629389172, 'p_value': 0.5585655964026822}, 'AUC': {'t_stat': -3.20136439051826, 'p_value': 0.004699791731238062}}, 'Two Shared vs Eleven Shared': {'Accuracy': {'t_stat': -1.212955167299299, 'p_value': 0.24001263795265013}, 'Sensitivity': {'t_stat': -2.545410713519758, 'p_value': 0.019750329433254346}, 'Specificity': {'t_stat': -1.0926006480900698, 'p_value': 0.28822976718638493}, 'AUC': {'t_stat': -5.211625558575092, 'p_value': 4.966441126682958e-05}}, 'Two Shared vs Twelve Shared': {'Accuracy': {'t_stat': -1.7446439013633592, 'p_value': 0.09720181565642984}, 'Sensitivity': {'t_stat': -1.7766517266008892, 'p_value': 0.09164077382368224}, 'Specificity': {'t_stat': -1.5549077730093082, 'p_value': 0.13646767282251762}, 'AUC': {'t_stat': -4.071135776648011, 'p_value': 0.000651290580314234}}, 'Two Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -1.1560523599895987, 'p_value': 0.2619900563817679}, 'Sensitivity': {'t_stat': -2.316660513924613, 'p_value': 0.03184510468115583}, 'Specificity': {'t_stat': -0.974464326336689, 'p_value': 0.34207022363691064}, 'AUC': {'t_stat': -3.8773862817163804, 'p_value': 0.0010138423205053933}}, 'Three Shared vs Four Shared': {'Accuracy': {'t_stat': -0.6139100400141969, 'p_value': 0.5465556484948557}, 'Sensitivity': {'t_stat': -0.26739018625764466, 'p_value': 0.792049709043837}, 'Specificity': {'t_stat': -0.640511060905219, 'p_value': 0.5294902163157313}, 'AUC': {'t_stat': -0.06091406111245627, 'p_value': 0.9520638420638997}}, 'Three Shared vs Five Shared': {'Accuracy': {'t_stat': -0.4682749312932572, 'p_value': 0.6449153090368457}, 'Sensitivity': {'t_stat': -0.1893512181922057, 'p_value': 0.8518256560947303}, 'Specificity': {'t_stat': -0.35684205639525535, 'p_value': 0.725142497507559}, 'AUC': {'t_stat': 0.04716936751447612, 'p_value': 0.962870521286989}}, 'Three Shared vs Six Shared': {'Accuracy': {'t_stat': -1.0947534402118964, 'p_value': 0.28730913753750054}, 'Sensitivity': {'t_stat': 0.272121591706364, 'p_value': 0.788463655575157}, 'Specificity': {'t_stat': -1.1694742225572443, 'p_value': 0.2566750513006057}, 'AUC': {'t_stat': 0.027436432406651137, 'p_value': 0.9783977703656397}}, 'Three Shared vs Seven Shared': {'Accuracy': {'t_stat': -0.7317796202187596, 'p_value': 0.4732332798083623}, 'Sensitivity': {'t_stat': -0.06554488998527588, 'p_value': 0.9484249190082769}, 'Specificity': {'t_stat': -0.6359308390424825, 'p_value': 0.5324076003206972}, 'AUC': {'t_stat': -0.1899471274064331, 'p_value': 0.8513651928125144}}, 'Three Shared vs Eight Shared': {'Accuracy': {'t_stat': -0.09826040536526194, 'p_value': 0.922754745302742}, 'Sensitivity': {'t_stat': -2.1127546397379264, 'p_value': 0.04808284668729398}, 'Specificity': {'t_stat': -0.016369246343253936, 'p_value': 0.987110485509056}, 'AUC': {'t_stat': -1.9808496597553373, 'p_value': 0.06227137528420346}}, 'Three Shared vs Nine Shared': {'Accuracy': {'t_stat': -2.55225055910513, 'p_value': 0.019465882290600527}, 'Sensitivity': {'t_stat': -0.6170243254423815, 'p_value': 0.5445426018552387}, 'Specificity': {'t_stat': -2.539869978203086, 'p_value': 0.019983618024455984}, 'AUC': {'t_stat': -2.589746169266834, 'p_value': 0.017973993833742236}}, 'Three Shared vs Ten Shared': {'Accuracy': {'t_stat': -0.9795566531990726, 'p_value': 0.3396140372381282}, 'Sensitivity': {'t_stat': -1.9763584988979526, 'p_value': 0.06281456186621394}, 'Specificity': {'t_stat': -0.9241592026520927, 'p_value': 0.3669923599365905}, 'AUC': {'t_stat': -2.7397421280160006, 'p_value': 0.013021218482642217}}, 'Three Shared vs Eleven Shared': {'Accuracy': {'t_stat': -1.5304318296569746, 'p_value': 0.1423897754759501}, 'Sensitivity': {'t_stat': -1.8761483688381697, 'p_value': 0.07608340298539505}, 'Specificity': {'t_stat': -1.4537244354170882, 'p_value': 0.16234195021773073}, 'AUC': {'t_stat': -4.50209679431886, 'p_value': 0.0002440207074025015}}, 'Three Shared vs Twelve Shared': {'Accuracy': {'t_stat': -1.9282120125842779, 'p_value': 0.06890859091108448}, 'Sensitivity': {'t_stat': -1.2276676482291504, 'p_value': 0.23456496561393933}, 'Specificity': {'t_stat': -1.8271556908102533, 'p_value': 0.08342790548237725}, 'AUC': {'t_stat': -2.957275961272011, 'p_value': 0.008090935242010421}}, 'Three Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -1.8680733758850057, 'p_value': 0.07725332681906043}, 'Sensitivity': {'t_stat': -1.854920206444243, 'p_value': 0.07919289134578665}, 'Specificity': {'t_stat': -1.6813385813704087, 'p_value': 0.10906189098127898}, 'AUC': {'t_stat': -3.4577470882787904, 'p_value': 0.0026362354111265102}}, 'Four Shared vs Five Shared': {'Accuracy': {'t_stat': 0.18465591430698053, 'p_value': 0.8554556471910453}, 'Sensitivity': {'t_stat': 0.054666820560171525, 'p_value': 0.9569746444946063}, 'Specificity': {'t_stat': 0.2846004059213465, 'p_value': 0.7790290498439056}, 'AUC': {'t_stat': 0.11789464740726778, 'p_value': 0.9073885598097324}}, 'Four Shared vs Six Shared': {'Accuracy': {'t_stat': -0.5973018426565917, 'p_value': 0.557357903036566}, 'Sensitivity': {'t_stat': 0.49084816838294987, 'p_value': 0.6291550659090912}, 'Specificity': {'t_stat': -0.6348315584031016, 'p_value': 0.5331090964487012}, 'AUC': {'t_stat': 0.11964729200732988, 'p_value': 0.9060186282523003}}, 'Four Shared vs Seven Shared': {'Accuracy': {'t_stat': -0.10070024523189984, 'p_value': 0.9208434489765258}, 'Sensitivity': {'t_stat': 0.18757736107052553, 'p_value': 0.8531966505663623}, 'Specificity': {'t_stat': 0.005408181437357876, 'p_value': 0.9957412973824254}, 'AUC': {'t_stat': -0.09635462337114102, 'p_value': 0.9242480129301992}}, 'Four Shared vs Eight Shared': {'Accuracy': {'t_stat': 0.5497374506167823, 'p_value': 0.5889037932290593}, 'Sensitivity': {'t_stat': -2.0709546169873896, 'p_value': 0.052226180933602234}, 'Specificity': {'t_stat': 0.6699677892074734, 'p_value': 0.5109395159443588}, 'AUC': {'t_stat': -1.4698719210229407, 'p_value': 0.15796197944635038}}, 'Four Shared vs Nine Shared': {'Accuracy': {'t_stat': -1.1213528964738328, 'p_value': 0.2761106673994181}, 'Sensitivity': {'t_stat': -0.36077113474987493, 'p_value': 0.7222505486744885}, 'Specificity': {'t_stat': -1.1503030066242743, 'p_value': 0.26429175516720843}, 'AUC': {'t_stat': -2.5082930455946904, 'p_value': 0.021363445790933907}}, 'Four Shared vs Ten Shared': {'Accuracy': {'t_stat': -0.26871161071044436, 'p_value': 0.7910476851162923}, 'Sensitivity': {'t_stat': -1.681671149342446, 'p_value': 0.10899648567391745}, 'Specificity': {'t_stat': -0.20603204238254355, 'p_value': 0.8389574883027819}, 'AUC': {'t_stat': -3.003326822335845, 'p_value': 0.007307690287138555}}, 'Four Shared vs Eleven Shared': {'Accuracy': {'t_stat': -0.779262247075849, 'p_value': 0.44542362033102056}, 'Sensitivity': {'t_stat': -2.0707308328180933, 'p_value': 0.05224920590422255}, 'Specificity': {'t_stat': -0.7435979468184086, 'p_value': 0.46621640556076194}, 'AUC': {'t_stat': -3.0321161664887955, 'p_value': 0.006855863795520288}}, 'Four Shared vs Twelve Shared': {'Accuracy': {'t_stat': -1.1799323433677886, 'p_value': 0.25259004723370004}, 'Sensitivity': {'t_stat': -0.9037765636481605, 'p_value': 0.377430623448769}, 'Specificity': {'t_stat': -1.0912884778757217, 'p_value': 0.2887919626682047}, 'AUC': {'t_stat': -2.599954177746095, 'p_value': 0.01758694449048535}}, 'Four Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -0.982832831625484, 'p_value': 0.3380403092313149}, 'Sensitivity': {'t_stat': -2.311468001012476, 'p_value': 0.03218651849885033}, 'Specificity': {'t_stat': -0.8940418405540559, 'p_value': 0.38248506493730505}, 'AUC': {'t_stat': -3.4496147916685547, 'p_value': 0.0026852671490922874}}, 'Five Shared vs Six Shared': {'Accuracy': {'t_stat': -0.8957319086485237, 'p_value': 0.38160434832895085}, 'Sensitivity': {'t_stat': 0.3688225293098043, 'p_value': 0.716337843295833}, 'Specificity': {'t_stat': -1.000483208572527, 'p_value': 0.32964903222541586}, 'AUC': {'t_stat': -0.02769017446946698, 'p_value': 0.9781980386185014}}, 'Five Shared vs Seven Shared': {'Accuracy': {'t_stat': -0.31103133306788, 'p_value': 0.7591634090280885}, 'Sensitivity': {'t_stat': 0.1027695497421732, 'p_value': 0.9192228054521263}, 'Specificity': {'t_stat': -0.3019477332997589, 'p_value': 0.7659721420868633}, 'AUC': {'t_stat': -0.19853050342188994, 'p_value': 0.8447389131897388}}, 'Five Shared vs Eight Shared': {'Accuracy': {'t_stat': 0.4319978504373698, 'p_value': 0.670605781023829}, 'Sensitivity': {'t_stat': -1.5506395158489221, 'p_value': 0.1374853834704529}, 'Specificity': {'t_stat': 0.39007405492357333, 'p_value': 0.7008202366099212}, 'AUC': {'t_stat': -1.5846585277316125, 'p_value': 0.12954722626576706}}, 'Five Shared vs Nine Shared': {'Accuracy': {'t_stat': -1.5996726875702327, 'p_value': 0.1261679449583292}, 'Sensitivity': {'t_stat': -0.37864812337981546, 'p_value': 0.7091470750332378}, 'Specificity': {'t_stat': -1.7377651084544545, 'p_value': 0.09843429503030388}, 'AUC': {'t_stat': -2.776175483196867, 'p_value': 0.012031508101008219}}, 'Five Shared vs Ten Shared': {'Accuracy': {'t_stat': -0.5780876767721888, 'p_value': 0.5699938513289309}, 'Sensitivity': {'t_stat': -1.4822149062761432, 'p_value': 0.1546795608841782}, 'Specificity': {'t_stat': -0.6073567156098832, 'p_value': 0.550804633145747}, 'AUC': {'t_stat': -2.396516088182181, 'p_value': 0.02699995118076275}}, 'Five Shared vs Eleven Shared': {'Accuracy': {'t_stat': -1.2146284784300876, 'p_value': 0.2393882356131425}, 'Sensitivity': {'t_stat': -1.4753040028032351, 'p_value': 0.15651044535113623}, 'Specificity': {'t_stat': -1.2087194974006916, 'p_value': 0.24159873563878098}, 'AUC': {'t_stat': -3.5794241691995348, 'p_value': 0.0019998907539203786}}, 'Five Shared vs Twelve Shared': {'Accuracy': {'t_stat': -2.379558324990652, 'p_value': 0.027967564845568976}, 'Sensitivity': {'t_stat': -0.9746093898592014, 'p_value': 0.34200008553353467}, 'Specificity': {'t_stat': -2.148418266349497, 'p_value': 0.04478599378647024}, 'AUC': {'t_stat': -3.7533513434169645, 'p_value': 0.0013456515542786542}}, 'Five Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -1.3260550566838927, 'p_value': 0.20054311545004969}, 'Sensitivity': {'t_stat': -1.564377594263346, 'p_value': 0.13423212995308514}, 'Specificity': {'t_stat': -1.2237055824710268, 'p_value': 0.23602263381978367}, 'AUC': {'t_stat': -4.425067620275265, 'p_value': 0.0002906715244730192}}, 'Six Shared vs Seven Shared': {'Accuracy': {'t_stat': 0.5961522592220692, 'p_value': 0.558109750927273}, 'Sensitivity': {'t_stat': -0.34277131933437444, 'p_value': 0.73553360782319}, 'Specificity': {'t_stat': 0.7868629861760306, 'p_value': 0.4410671166282558}, 'AUC': {'t_stat': -0.18275762994018158, 'p_value': 0.856924175466692}}, 'Six Shared vs Eight Shared': {'Accuracy': {'t_stat': 1.1948943154551557, 'p_value': 0.24683119802779768}, 'Sensitivity': {'t_stat': -2.0565799374352047, 'p_value': 0.053723921333641}, 'Specificity': {'t_stat': 1.3411302312291031, 'p_value': 0.19569140777460056}, 'AUC': {'t_stat': -1.6724548991659316, 'p_value': 0.11082146798306339}}, 'Six Shared vs Nine Shared': {'Accuracy': {'t_stat': -0.4244365590373926, 'p_value': 0.6760146989970708}, 'Sensitivity': {'t_stat': -0.724697854482748, 'p_value': 0.4774678496228585}, 'Specificity': {'t_stat': -0.3620511039968022, 'p_value': 0.721309366244977}, 'AUC': {'t_stat': -4.220379801252231, 'p_value': 0.0004632718374417377}}, 'Six Shared vs Ten Shared': {'Accuracy': {'t_stat': 0.4160281775116482, 'p_value': 0.6820508573847129}, 'Sensitivity': {'t_stat': -1.643970183997459, 'p_value': 0.11662742401671425}, 'Specificity': {'t_stat': 0.5399101855912778, 'p_value': 0.5955310742645487}, 'AUC': {'t_stat': -3.4532623629791384, 'p_value': 0.0026631648199707974}}, 'Six Shared vs Eleven Shared': {'Accuracy': {'t_stat': -0.014689993020496917, 'p_value': 0.9884326616850251}, 'Sensitivity': {'t_stat': -2.033029773868273, 'p_value': 0.05626150136676125}, 'Specificity': {'t_stat': 0.11006117418254416, 'p_value': 0.9135150690697929}, 'AUC': {'t_stat': -3.963806908506531, 'p_value': 0.0008322301149870918}}, 'Six Shared vs Twelve Shared': {'Accuracy': {'t_stat': -0.9209824462452423, 'p_value': 0.3686063258328597}, 'Sensitivity': {'t_stat': -1.6459811108059206, 'p_value': 0.1162092339406891}, 'Specificity': {'t_stat': -0.6664063555756601, 'p_value': 0.5131627253613051}, 'AUC': {'t_stat': -3.76911890736021, 'p_value': 0.001298104474802821}}, 'Six Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -0.11303623171645774, 'p_value': 0.9111876197869168}, 'Sensitivity': {'t_stat': -1.8154365326675728, 'p_value': 0.08527412941118771}, 'Specificity': {'t_stat': 0.052783674303312246, 'p_value': 0.9584552950084362}, 'AUC': {'t_stat': -4.426671337024123, 'p_value': 0.0002896138952939582}}, 'Seven Shared vs Eight Shared': {'Accuracy': {'t_stat': 0.7620713802866719, 'p_value': 0.45537412995888893}, 'Sensitivity': {'t_stat': -1.9765581599807192, 'p_value': 0.0627903241506164}, 'Specificity': {'t_stat': 0.73969530327968, 'p_value': 0.4685265899094444}, 'AUC': {'t_stat': -1.6993745416429589, 'p_value': 0.10556286241731729}}, 'Seven Shared vs Nine Shared': {'Accuracy': {'t_stat': -0.9250835895360323, 'p_value': 0.3665236160418134}, 'Sensitivity': {'t_stat': -0.47130347146559753, 'p_value': 0.642790572534253}, 'Specificity': {'t_stat': -1.022539936000663, 'p_value': 0.31936931350277153}, 'AUC': {'t_stat': -1.8367582354210457, 'p_value': 0.08194128477400651}}, 'Seven Shared vs Ten Shared': {'Accuracy': {'t_stat': -0.16879970066543484, 'p_value': 0.8677382379043654}, 'Sensitivity': {'t_stat': -1.8513986775087299, 'p_value': 0.07971938140749926}, 'Specificity': {'t_stat': -0.22293024364803188, 'p_value': 0.8259689180526919}, 'AUC': {'t_stat': -2.2439847448536234, 'p_value': 0.03694214050612782}}, 'Seven Shared vs Eleven Shared': {'Accuracy': {'t_stat': -0.5859501149098918, 'p_value': 0.5648054001108288}, 'Sensitivity': {'t_stat': -2.1572245150739753, 'p_value': 0.04400424128821853}, 'Specificity': {'t_stat': -0.6633793339769058, 'p_value': 0.5150566030092436}, 'AUC': {'t_stat': -4.294813548339738, 'p_value': 0.0003909704841182547}}, 'Seven Shared vs Twelve Shared': {'Accuracy': {'t_stat': -1.0657824632867037, 'p_value': 0.2998787857729321}, 'Sensitivity': {'t_stat': -0.8642067216354581, 'p_value': 0.39825373206906145}, 'Specificity': {'t_stat': -1.0912680871428513, 'p_value': 0.2888007053138278}, 'AUC': {'t_stat': -2.2596597342750107, 'p_value': 0.03578310386019676}}, 'Seven Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -0.7458716532877708, 'p_value': 0.464873632409574}, 'Sensitivity': {'t_stat': -2.087347465045573, 'p_value': 0.05056425445402856}, 'Specificity': {'t_stat': -0.7502504962656301, 'p_value': 0.46229419569668406}, 'AUC': {'t_stat': -3.6081918609580836, 'p_value': 0.0018732034971743322}}, 'Eight Shared vs Nine Shared': {'Accuracy': {'t_stat': -2.0863356550803656, 'p_value': 0.050665431442613464}, 'Sensitivity': {'t_stat': 1.3796898392451178, 'p_value': 0.1837043699335041}, 'Specificity': {'t_stat': -2.1261446891958893, 'p_value': 0.0468200220325128}, 'AUC': {'t_stat': -0.8888744819096428, 'p_value': 0.38518618644437264}}, 'Eight Shared vs Ten Shared': {'Accuracy': {'t_stat': -1.0677984103841842, 'p_value': 0.29899148458930946}, 'Sensitivity': {'t_stat': -0.10802577591155602, 'p_value': 0.915107866640944}, 'Specificity': {'t_stat': -1.0698646849521622, 'p_value': 0.2980839984558259}, 'AUC': {'t_stat': -1.462542471064285, 'p_value': 0.1599379554102681}}, 'Eight Shared vs Eleven Shared': {'Accuracy': {'t_stat': -1.5148570195958162, 'p_value': 0.14626803927460136}, 'Sensitivity': {'t_stat': -0.3113839530961981, 'p_value': 0.7588995002353858}, 'Specificity': {'t_stat': -1.5255647547864855, 'p_value': 0.14359245924983524}, 'AUC': {'t_stat': -2.8078270391735973, 'p_value': 0.011230474217470848}}, 'Eight Shared vs Twelve Shared': {'Accuracy': {'t_stat': -2.3306925881900336, 'p_value': 0.030939177978176077}, 'Sensitivity': {'t_stat': 0.5849311287260717, 'p_value': 0.5654764464003225}, 'Specificity': {'t_stat': -2.349064099627134, 'p_value': 0.0297891820269309}, 'AUC': {'t_stat': -1.6804970407154851, 'p_value': 0.10922754441877026}}, 'Eight Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -1.8044102916385143, 'p_value': 0.08704366238432959}, 'Sensitivity': {'t_stat': -0.1771515934478895, 'p_value': 0.861264156251811}, 'Specificity': {'t_stat': -1.8422513451765345, 'p_value': 0.08110134173272318}, 'AUC': {'t_stat': -2.276488480363213, 'p_value': 0.034576094389611484}}, 'Nine Shared vs Ten Shared': {'Accuracy': {'t_stat': 1.0520003847076935, 'p_value': 0.3059956662752235}, 'Sensitivity': {'t_stat': -1.0327258440152776, 'p_value': 0.3146993360958246}, 'Specificity': {'t_stat': 1.1235731601874581, 'p_value': 0.27519066081611065}, 'AUC': {'t_stat': -0.2806778308559335, 'p_value': 0.7819910003659095}}, 'Nine Shared vs Eleven Shared': {'Accuracy': {'t_stat': 0.6610744436059245, 'p_value': 0.5165013011122505}, 'Sensitivity': {'t_stat': -1.6999681455691047, 'p_value': 0.10544935476513842}, 'Specificity': {'t_stat': 0.7642370749794591, 'p_value': 0.4541131698227757}, 'AUC': {'t_stat': -1.66265795488645, 'p_value': 0.11278994627192508}}, 'Nine Shared vs Twelve Shared': {'Accuracy': {'t_stat': -0.24683575301512448, 'p_value': 0.8076827015657154}, 'Sensitivity': {'t_stat': -0.6420020545152586, 'p_value': 0.5285424222496978}, 'Specificity': {'t_stat': -0.09823175546846045, 'p_value': 0.9227771916475136}, 'AUC': {'t_stat': -0.7565760674702168, 'p_value': 0.45858327944382216}}, 'Nine Shared vs Thirteen Shared': {'Accuracy': {'t_stat': 0.47296769855744236, 'p_value': 0.6416243402281177}, 'Sensitivity': {'t_stat': -1.4777453712516562, 'p_value': 0.15586164281928822}, 'Specificity': {'t_stat': 0.6186037998796462, 'p_value': 0.5435231669488996}, 'AUC': {'t_stat': -1.374536522725016, 'p_value': 0.18527158422961776}}, 'Ten Shared vs Eleven Shared': {'Accuracy': {'t_stat': -0.5656556142542417, 'p_value': 0.5782476305375004}, 'Sensitivity': {'t_stat': -0.2066688135962631, 'p_value': 0.8384671568949081}, 'Specificity': {'t_stat': -0.5861010935920337, 'p_value': 0.5647060093617884}, 'AUC': {'t_stat': -1.5667689948471262, 'p_value': 0.13367245309339673}}, 'Ten Shared vs Twelve Shared': {'Accuracy': {'t_stat': -1.295287502194276, 'p_value': 0.21073893481258996}, 'Sensitivity': {'t_stat': 0.5327596332632446, 'p_value': 0.6003762431750379}, 'Specificity': {'t_stat': -1.2558401971075734, 'p_value': 0.22439821249283726}, 'AUC': {'t_stat': -0.5325040888838877, 'p_value': 0.6005497544537937}}, 'Ten Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -0.6830065256334795, 'p_value': 0.5028467350092496}, 'Sensitivity': {'t_stat': -0.04127736224259473, 'p_value': 0.9675054574017394}, 'Specificity': {'t_stat': -0.6066374443683719, 'p_value': 0.5512720553523011}, 'AUC': {'t_stat': -1.0330400232279409, 'p_value': 0.31455606772599065}}, 'Eleven Shared vs Twelve Shared': {'Accuracy': {'t_stat': -0.8191895890156846, 'p_value': 0.4228352314232994}, 'Sensitivity': {'t_stat': 0.7537934330728076, 'p_value': 0.4602134936619522}, 'Specificity': {'t_stat': -0.7105023712467945, 'p_value': 0.486023213809082}, 'AUC': {'t_stat': 1.0931130415993653, 'p_value': 0.28801045027380634}}, 'Eleven Shared vs Thirteen Shared': {'Accuracy': {'t_stat': -0.12536224400970156, 'p_value': 0.9015537329025567}, 'Sensitivity': {'t_stat': 0.17533188770051022, 'p_value': 0.8626738843565722}, 'Specificity': {'t_stat': -0.05514811632257748, 'p_value': 0.956596244379184}, 'AUC': {'t_stat': 0.7334497101510203, 'p_value': 0.4722379049422145}}, 'Twelve Shared vs Thirteen Shared': {'Accuracy': {'t_stat': 0.6403237446899046, 'p_value': 0.5296093553942861}, 'Sensitivity': {'t_stat': -0.5771435621623928, 'p_value': 0.5706185243870536}, 'Specificity': {'t_stat': 0.6407436824738189, 'p_value': 0.5293422820912136}, 'AUC': {'t_stat': -0.523006234087276, 'p_value': 0.6070159751814628}}}\n"
     ]
    }
   ],
   "source": [
    "print(t_test_result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Two Shared vs Three Shared\n",
      "{'t_stat': 0.4718886775693806, 'p_value': 0.6423803718375879}\n",
      "Two Shared vs Four Shared\n",
      "{'t_stat': -0.22105680368423875, 'p_value': 0.827406451989789}\n",
      "Two Shared vs Five Shared\n",
      "{'t_stat': -0.031589774029780936, 'p_value': 0.9751286830114254}\n",
      "Two Shared vs Six Shared\n",
      "{'t_stat': -1.1097725596972379, 'p_value': 0.2809459296218739}\n",
      "Two Shared vs Seven Shared\n",
      "{'t_stat': -0.4560457772804352, 'p_value': 0.6535266919293146}\n",
      "Two Shared vs Eight Shared\n",
      "{'t_stat': 0.34932346211127724, 'p_value': 0.7306882580734704}\n",
      "Two Shared vs Nine Shared\n",
      "{'t_stat': -1.5186655749608875, 'p_value': 0.1453116962094483}\n",
      "Two Shared vs Ten Shared\n",
      "{'t_stat': -0.7298273915688733, 'p_value': 0.474398389244731}\n",
      "Two Shared vs Eleven Shared\n",
      "{'t_stat': -1.212955167299299, 'p_value': 0.24001263795265013}\n",
      "Two Shared vs Twelve Shared\n",
      "{'t_stat': -1.7446439013633592, 'p_value': 0.09720181565642984}\n",
      "Two Shared vs Thirteen Shared\n",
      "{'t_stat': -1.1560523599895987, 'p_value': 0.2619900563817679}\n",
      "Three Shared vs Four Shared\n",
      "{'t_stat': -0.6139100400141969, 'p_value': 0.5465556484948557}\n",
      "Three Shared vs Five Shared\n",
      "{'t_stat': -0.4682749312932572, 'p_value': 0.6449153090368457}\n",
      "Three Shared vs Six Shared\n",
      "{'t_stat': -1.0947534402118964, 'p_value': 0.28730913753750054}\n",
      "Three Shared vs Seven Shared\n",
      "{'t_stat': -0.7317796202187596, 'p_value': 0.4732332798083623}\n",
      "Three Shared vs Eight Shared\n",
      "{'t_stat': -0.09826040536526194, 'p_value': 0.922754745302742}\n",
      "Three Shared vs Nine Shared\n",
      "{'t_stat': -2.55225055910513, 'p_value': 0.019465882290600527}\n",
      "Three Shared vs Ten Shared\n",
      "{'t_stat': -0.9795566531990726, 'p_value': 0.3396140372381282}\n",
      "Three Shared vs Eleven Shared\n",
      "{'t_stat': -1.5304318296569746, 'p_value': 0.1423897754759501}\n",
      "Three Shared vs Twelve Shared\n",
      "{'t_stat': -1.9282120125842779, 'p_value': 0.06890859091108448}\n",
      "Three Shared vs Thirteen Shared\n",
      "{'t_stat': -1.8680733758850057, 'p_value': 0.07725332681906043}\n",
      "Four Shared vs Five Shared\n",
      "{'t_stat': 0.18465591430698053, 'p_value': 0.8554556471910453}\n",
      "Four Shared vs Six Shared\n",
      "{'t_stat': -0.5973018426565917, 'p_value': 0.557357903036566}\n",
      "Four Shared vs Seven Shared\n",
      "{'t_stat': -0.10070024523189984, 'p_value': 0.9208434489765258}\n",
      "Four Shared vs Eight Shared\n",
      "{'t_stat': 0.5497374506167823, 'p_value': 0.5889037932290593}\n",
      "Four Shared vs Nine Shared\n",
      "{'t_stat': -1.1213528964738328, 'p_value': 0.2761106673994181}\n",
      "Four Shared vs Ten Shared\n",
      "{'t_stat': -0.26871161071044436, 'p_value': 0.7910476851162923}\n",
      "Four Shared vs Eleven Shared\n",
      "{'t_stat': -0.779262247075849, 'p_value': 0.44542362033102056}\n",
      "Four Shared vs Twelve Shared\n",
      "{'t_stat': -1.1799323433677886, 'p_value': 0.25259004723370004}\n",
      "Four Shared vs Thirteen Shared\n",
      "{'t_stat': -0.982832831625484, 'p_value': 0.3380403092313149}\n",
      "Five Shared vs Six Shared\n",
      "{'t_stat': -0.8957319086485237, 'p_value': 0.38160434832895085}\n",
      "Five Shared vs Seven Shared\n",
      "{'t_stat': -0.31103133306788, 'p_value': 0.7591634090280885}\n",
      "Five Shared vs Eight Shared\n",
      "{'t_stat': 0.4319978504373698, 'p_value': 0.670605781023829}\n",
      "Five Shared vs Nine Shared\n",
      "{'t_stat': -1.5996726875702327, 'p_value': 0.1261679449583292}\n",
      "Five Shared vs Ten Shared\n",
      "{'t_stat': -0.5780876767721888, 'p_value': 0.5699938513289309}\n",
      "Five Shared vs Eleven Shared\n",
      "{'t_stat': -1.2146284784300876, 'p_value': 0.2393882356131425}\n",
      "Five Shared vs Twelve Shared\n",
      "{'t_stat': -2.379558324990652, 'p_value': 0.027967564845568976}\n",
      "Five Shared vs Thirteen Shared\n",
      "{'t_stat': -1.3260550566838927, 'p_value': 0.20054311545004969}\n",
      "Six Shared vs Seven Shared\n",
      "{'t_stat': 0.5961522592220692, 'p_value': 0.558109750927273}\n",
      "Six Shared vs Eight Shared\n",
      "{'t_stat': 1.1948943154551557, 'p_value': 0.24683119802779768}\n",
      "Six Shared vs Nine Shared\n",
      "{'t_stat': -0.4244365590373926, 'p_value': 0.6760146989970708}\n",
      "Six Shared vs Ten Shared\n",
      "{'t_stat': 0.4160281775116482, 'p_value': 0.6820508573847129}\n",
      "Six Shared vs Eleven Shared\n",
      "{'t_stat': -0.014689993020496917, 'p_value': 0.9884326616850251}\n",
      "Six Shared vs Twelve Shared\n",
      "{'t_stat': -0.9209824462452423, 'p_value': 0.3686063258328597}\n",
      "Six Shared vs Thirteen Shared\n",
      "{'t_stat': -0.11303623171645774, 'p_value': 0.9111876197869168}\n",
      "Seven Shared vs Eight Shared\n",
      "{'t_stat': 0.7620713802866719, 'p_value': 0.45537412995888893}\n",
      "Seven Shared vs Nine Shared\n",
      "{'t_stat': -0.9250835895360323, 'p_value': 0.3665236160418134}\n",
      "Seven Shared vs Ten Shared\n",
      "{'t_stat': -0.16879970066543484, 'p_value': 0.8677382379043654}\n",
      "Seven Shared vs Eleven Shared\n",
      "{'t_stat': -0.5859501149098918, 'p_value': 0.5648054001108288}\n",
      "Seven Shared vs Twelve Shared\n",
      "{'t_stat': -1.0657824632867037, 'p_value': 0.2998787857729321}\n",
      "Seven Shared vs Thirteen Shared\n",
      "{'t_stat': -0.7458716532877708, 'p_value': 0.464873632409574}\n",
      "Eight Shared vs Nine Shared\n",
      "{'t_stat': -2.0863356550803656, 'p_value': 0.050665431442613464}\n",
      "Eight Shared vs Ten Shared\n",
      "{'t_stat': -1.0677984103841842, 'p_value': 0.29899148458930946}\n",
      "Eight Shared vs Eleven Shared\n",
      "{'t_stat': -1.5148570195958162, 'p_value': 0.14626803927460136}\n",
      "Eight Shared vs Twelve Shared\n",
      "{'t_stat': -2.3306925881900336, 'p_value': 0.030939177978176077}\n",
      "Eight Shared vs Thirteen Shared\n",
      "{'t_stat': -1.8044102916385143, 'p_value': 0.08704366238432959}\n",
      "Nine Shared vs Ten Shared\n",
      "{'t_stat': 1.0520003847076935, 'p_value': 0.3059956662752235}\n",
      "Nine Shared vs Eleven Shared\n",
      "{'t_stat': 0.6610744436059245, 'p_value': 0.5165013011122505}\n",
      "Nine Shared vs Twelve Shared\n",
      "{'t_stat': -0.24683575301512448, 'p_value': 0.8076827015657154}\n",
      "Nine Shared vs Thirteen Shared\n",
      "{'t_stat': 0.47296769855744236, 'p_value': 0.6416243402281177}\n",
      "Ten Shared vs Eleven Shared\n",
      "{'t_stat': -0.5656556142542417, 'p_value': 0.5782476305375004}\n",
      "Ten Shared vs Twelve Shared\n",
      "{'t_stat': -1.295287502194276, 'p_value': 0.21073893481258996}\n",
      "Ten Shared vs Thirteen Shared\n",
      "{'t_stat': -0.6830065256334795, 'p_value': 0.5028467350092496}\n",
      "Eleven Shared vs Twelve Shared\n",
      "{'t_stat': -0.8191895890156846, 'p_value': 0.4228352314232994}\n",
      "Eleven Shared vs Thirteen Shared\n",
      "{'t_stat': -0.12536224400970156, 'p_value': 0.9015537329025567}\n",
      "Twelve Shared vs Thirteen Shared\n",
      "{'t_stat': 0.6403237446899046, 'p_value': 0.5296093553942861}\n",
      "Data successfully exported\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import openpyxl\n",
    "\n",
    "# Create a list of rows, where each row is a dict that will become one row in the final DataFrame\n",
    "rows = []\n",
    "\n",
    "for shared_label, metrics_dict in t_test_result.items():\n",
    "    print(shared_label)\n",
    "    print(metrics_dict['Accuracy'])\n",
    "    # Each \"metrics_dict\" has keys like 'Accuracy', 'Sensitivity', 'Specificity', 'AUC'.\n",
    "    # Each of these keys corresponds to a list of 20 floats (in your example).\n",
    "    for metric_name, metric_values in metrics_dict.items():\n",
    "        # metric_values should be {\"t_stat\": X, \"p_value\": Y}\n",
    "        row = {\n",
    "            \"Shared Label\": shared_label,\n",
    "            \"Metric\": metric_name,\n",
    "            \"t_stat\": metric_values[\"t_stat\"],\n",
    "            \"p_value\": metric_values[\"p_value\"]\n",
    "        }\n",
    "        rows.append(row)\n",
    "# Convert the list of dicts into a pandas DataFrame\n",
    "df = pd.DataFrame(rows)\n",
    "\n",
    "# Now export this DataFrame to Excel\n",
    "df.to_excel(r\"data\\t-tests\\Experiment 3 Result.xlsx\", index=False)\n",
    "\n",
    "print(\"Data successfully exported\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Downloading Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import nbformat\n",
    "\n",
    "# Load the notebook\n",
    "with open('FL_Embedding_Classifier.ipynb', 'r', encoding='utf-8') as f:\n",
    "    notebook = nbformat.read(f, as_version=4)\n",
    "\n",
    "# Create a directory to save images\n",
    "if not os.path.exists('downloaded_images'):\n",
    "    os.makedirs('downloaded_images')\n",
    "\n",
    "# Get all image objects from the notebook\n",
    "images = [cell for cell in notebook.cells if 'outputs' in cell and cell['outputs'] \n",
    "          and any('image/png' in output.get('data', {}) for output in cell['outputs'])]\n",
    "\n",
    "# Save each image\n",
    "for i, cell in enumerate(images):\n",
    "    for output in cell['outputs']:\n",
    "        if 'image/png' in output.get('data', {}):\n",
    "            image_data = output['data']['image/png']\n",
    "            with open(f'downloaded_images/image_{i}.png', 'wb') as f:\n",
    "                import base64\n",
    "                f.write(base64.b64decode(image_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "gpuType": "T4",
   "provenance": []
  },
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.8"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "04ff5f37db7c42f2980f807bc3032586": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_74a45523ed954b0abd9e7a05045d2daf",
      "placeholder": "​",
      "style": "IPY_MODEL_c2058ed868ff4815a9ef7ea89e85f022",
      "value": " 10.7k/10.7k [00:00&lt;00:00, 236kB/s]"
     }
    },
    "0a66bb0689824b83a0230934c12374e3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "0cdbc7260ad346048a87a3f8ba22486e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_bf354c1bbf524f97a07cd2a6c854e22c",
      "placeholder": "​",
      "style": "IPY_MODEL_199a73af56a5402e9ad15644a7c72755",
      "value": " 349/349 [00:00&lt;00:00, 8.26kB/s]"
     }
    },
    "0ea78511c0d54dd1acbd29baaf0cbf58": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_abdbb22939fd41db9b0a3ee58c7ed13b",
       "IPY_MODEL_c92270c929c24bcb847c6739686a799b",
       "IPY_MODEL_7b87dd0653c14863a8a2af560d5e7c42"
      ],
      "layout": "IPY_MODEL_fec6c46a94314fff8f5fa6db76c722a2"
     }
    },
    "10c6a410b1ca4210812a84b8d1bd9984": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "12e62aa39c8a475ebd65aaf5e5e84338": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_89c8e756392047b48e155bc153b4f6dd",
      "placeholder": "​",
      "style": "IPY_MODEL_9ce5c894beb448b1b5935907c38d9138",
      "value": "tokenizer_config.json: 100%"
     }
    },
    "14261702d08a4d019e8ce8ee1f08d4ce": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_12e62aa39c8a475ebd65aaf5e5e84338",
       "IPY_MODEL_22223361bfd3496ca312f53e3653cfed",
       "IPY_MODEL_224ab887d94a455cb3664d0c7d7ddde2"
      ],
      "layout": "IPY_MODEL_81ae5aa2028c46248b01d916cce292fa"
     }
    },
    "14822fe23d064263bafda0ecaa573e15": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "14cb9e333a89414b9167d21d26760cff": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "16d38234182b49dda5d3115621aecb39": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "199a73af56a5402e9ad15644a7c72755": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "1bc226c51f954bbf98c13d2310940135": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1cc1e2d9840149c497d7d90b5dbb2d55": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "1e89725dd21a4a28a127279a274819d8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2153807b370049bd95c515e53f487f72": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "22223361bfd3496ca312f53e3653cfed": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_fbb3251aed0743e1be86088735d2ba33",
      "max": 350,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_22617e8f1b384d17b2481d77edb94658",
      "value": 350
     }
    },
    "224ab887d94a455cb3664d0c7d7ddde2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_a02025fa486a4cab95eb79496bc3e7ea",
      "placeholder": "​",
      "style": "IPY_MODEL_89e43961094244889ca5e14889d9f815",
      "value": " 350/350 [00:00&lt;00:00, 11.5kB/s]"
     }
    },
    "22617e8f1b384d17b2481d77edb94658": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "2319ac540cdb44eba7715740e1f391b3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "2377c5a7810c4e86849f2f1889cb0a5e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "25d29337a9544ee98ba83c7e9d75f853": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "267f0cdfde1f42cfb5a09d0da0c453ea": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "280332c72db54753a188f6abf68eaaa9": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2803d2f51c944e789952f5e4368c6c06": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "291e805751f84bb3a29d25c8b7cb3204": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "2c7d422cafe04b1cb2112f28fd631194": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_6b07c11c0b634256be6c9dc37369fbb5",
      "placeholder": "​",
      "style": "IPY_MODEL_bf34bae971834a2ca8ab1ac10e833986",
      "value": "1_Pooling/config.json: 100%"
     }
    },
    "2d273cb8c27347b68a191ee17dc89311": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "2d46775c61384019ad3479eab413e7a9": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2d94a369fc234cbc8452df68917e75b6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5e91ad19f8ad463f9cb5a0f9d1e8c294",
      "max": 10659,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_1cc1e2d9840149c497d7d90b5dbb2d55",
      "value": 10659
     }
    },
    "2fbd5496ad3944fd92a3c28f8d3e31a1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_76496dac60c44c839bb441f2f4ff6f9f",
      "placeholder": "​",
      "style": "IPY_MODEL_57453cbecc0b4d7ea786935449271c90",
      "value": "config_sentence_transformers.json: 100%"
     }
    },
    "30a058a01b434739adf517283900f1c8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "332a06b57849496c9a9b73b0cf2b4737": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_2c7d422cafe04b1cb2112f28fd631194",
       "IPY_MODEL_56762b63b5c74f56bdeb19012c7e509b",
       "IPY_MODEL_fe74971a69824ace86c03a408afe7100"
      ],
      "layout": "IPY_MODEL_af86bd6b8bef45c99bc6d61af2c4f71f"
     }
    },
    "33c5f2daa3a146e4ad66a753a4969e71": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "35750f3fe46e40b99058d14ca7ed800b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_fffef1e3409c4e929c21060b07853091",
      "placeholder": "​",
      "style": "IPY_MODEL_692fbf8b0b7542a39c3978569db3a9f8",
      "value": "model.safetensors: 100%"
     }
    },
    "3960d078abed4760b043dd20fddcf949": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2803d2f51c944e789952f5e4368c6c06",
      "placeholder": "​",
      "style": "IPY_MODEL_267f0cdfde1f42cfb5a09d0da0c453ea",
      "value": " 612/612 [00:00&lt;00:00, 13.8kB/s]"
     }
    },
    "3987a825b2584353a2be546f63911a0c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_4f3486a446af44019511654af1f51312",
      "placeholder": "​",
      "style": "IPY_MODEL_14cb9e333a89414b9167d21d26760cff",
      "value": " 90.9M/90.9M [00:00&lt;00:00, 122MB/s]"
     }
    },
    "3f2a6b57c81c4dc99404363b0b0e5632": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_1e89725dd21a4a28a127279a274819d8",
      "placeholder": "​",
      "style": "IPY_MODEL_0a66bb0689824b83a0230934c12374e3",
      "value": " 232k/232k [00:00&lt;00:00, 1.83MB/s]"
     }
    },
    "412ec4ec63d448939556ff9f40e5805b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_a53e2fc70d75494da4d6106f6608875b",
      "placeholder": "​",
      "style": "IPY_MODEL_77d4965ca1c1401381aca48ec309280e",
      "value": "README.md: 100%"
     }
    },
    "41be3437b15c4e57a75534d238a829a5": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4a8445cda576479ab8663568b2e58fe0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "4b8aa7e353f045dd8297de8c688b9d40": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2d46775c61384019ad3479eab413e7a9",
      "max": 349,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_25d29337a9544ee98ba83c7e9d75f853",
      "value": 349
     }
    },
    "4f3486a446af44019511654af1f51312": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "50b11ecfa6ee4d4d8d4fc97a31a4551b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "52067038eb414f5f983281696bcf84b1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_658c15a0430b4a788e9874b236cfb86d",
      "max": 612,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_701a5fc9c6ac4efebee807273eae66e1",
      "value": 612
     }
    },
    "56762b63b5c74f56bdeb19012c7e509b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b4717e6adc3e42648ffbf1e202a852ca",
      "max": 190,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_71cef30c369b4878823b8a4db5b44f5b",
      "value": 190
     }
    },
    "57453cbecc0b4d7ea786935449271c90": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "590a4e7454af492a8910e8fc1940b872": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_b73272664dfb4605b1d3a85928865304",
       "IPY_MODEL_52067038eb414f5f983281696bcf84b1",
       "IPY_MODEL_3960d078abed4760b043dd20fddcf949"
      ],
      "layout": "IPY_MODEL_14822fe23d064263bafda0ecaa573e15"
     }
    },
    "59d57ba6032a46d79628065ec01cff0f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "59f57f5ddfd24b53a4f1621f1a6389bb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5e61f9e947714e64a29e116c44a124f0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e48c72f5aa414987a2ce9409e9b5ad69",
      "placeholder": "​",
      "style": "IPY_MODEL_2d273cb8c27347b68a191ee17dc89311",
      "value": "sentence_bert_config.json: 100%"
     }
    },
    "5e91ad19f8ad463f9cb5a0f9d1e8c294": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "60bf8b23189b45949746bed8dbb91765": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "6274cc6165234aaba2c930d23350b362": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_5e61f9e947714e64a29e116c44a124f0",
       "IPY_MODEL_e2c7cbcbf7cd4e0ca1352925b8c7bdea",
       "IPY_MODEL_6b70db174de04c9ba25282525de05387"
      ],
      "layout": "IPY_MODEL_41be3437b15c4e57a75534d238a829a5"
     }
    },
    "63dc8025965242e09dfb7fffb70c953a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_280332c72db54753a188f6abf68eaaa9",
      "placeholder": "​",
      "style": "IPY_MODEL_f7d8f5d7761541ec95481008339f2420",
      "value": "modules.json: 100%"
     }
    },
    "64a3c90a4ba94283bbc717031722190c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "658c15a0430b4a788e9874b236cfb86d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "692fbf8b0b7542a39c3978569db3a9f8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "6b07c11c0b634256be6c9dc37369fbb5": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "6b70db174de04c9ba25282525de05387": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_8cd133cc3bd44a9cb49c7d1b6a4db642",
      "placeholder": "​",
      "style": "IPY_MODEL_16d38234182b49dda5d3115621aecb39",
      "value": " 53.0/53.0 [00:00&lt;00:00, 1.67kB/s]"
     }
    },
    "6bdcd35840594df4ab40d4c64e660dba": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "701a5fc9c6ac4efebee807273eae66e1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "71cef30c369b4878823b8a4db5b44f5b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "71f5342382cb4fdf818c03978a4ccccb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_fb9ef2497d3b47a297501004b0176cde",
       "IPY_MODEL_b63ec0fd33c448e89cc03e6000abee00",
       "IPY_MODEL_7da0f13278624d2a91348680b4f9e6be"
      ],
      "layout": "IPY_MODEL_8a82c40aedbc4555a03a8c29981740dd"
     }
    },
    "74a45523ed954b0abd9e7a05045d2daf": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "76496dac60c44c839bb441f2f4ff6f9f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "77d4965ca1c1401381aca48ec309280e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "7a839c9bc6ca43fdbf6d63cfe823af63": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7b87dd0653c14863a8a2af560d5e7c42": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b5a63beb6d2b4249871961e1a0a9b937",
      "placeholder": "​",
      "style": "IPY_MODEL_f94f4f97ce194f77bdd6d4bdfea4edea",
      "value": " 112/112 [00:00&lt;00:00, 1.66kB/s]"
     }
    },
    "7da0f13278624d2a91348680b4f9e6be": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_10c6a410b1ca4210812a84b8d1bd9984",
      "placeholder": "​",
      "style": "IPY_MODEL_c2d096efb1db4dde86fd42ff3a4df605",
      "value": " 466k/466k [00:00&lt;00:00, 3.66MB/s]"
     }
    },
    "81ae5aa2028c46248b01d916cce292fa": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "86b9a304e58446939dbb1d7a023de81f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_87034250c3894eee883cb3ecb9347078",
       "IPY_MODEL_efa59884be0e4fa6bb6ddf4d4885ab52",
       "IPY_MODEL_3f2a6b57c81c4dc99404363b0b0e5632"
      ],
      "layout": "IPY_MODEL_60bf8b23189b45949746bed8dbb91765"
     }
    },
    "87034250c3894eee883cb3ecb9347078": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ab01960dd6954af78441736c8669b241",
      "placeholder": "​",
      "style": "IPY_MODEL_e3f45f680ec84c4eb6ee402b187e38a8",
      "value": "vocab.txt: 100%"
     }
    },
    "874a74254ebc4d71a119f67e137c0551": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_2fbd5496ad3944fd92a3c28f8d3e31a1",
       "IPY_MODEL_a236061375e947dc9612c24adbe4ff0b",
       "IPY_MODEL_cd89e860240142e28da826f6f49ba240"
      ],
      "layout": "IPY_MODEL_9c8aa38595dc4b7e8ca640d03cc19694"
     }
    },
    "879a060bd57146398f7c175704456532": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "887871630cd944d89907336e3124835e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_412ec4ec63d448939556ff9f40e5805b",
       "IPY_MODEL_2d94a369fc234cbc8452df68917e75b6",
       "IPY_MODEL_04ff5f37db7c42f2980f807bc3032586"
      ],
      "layout": "IPY_MODEL_d357f7f4bd3f4af7af517dd7a725d2f0"
     }
    },
    "89c8e756392047b48e155bc153b4f6dd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "89e43961094244889ca5e14889d9f815": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "8a82c40aedbc4555a03a8c29981740dd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "8cd133cc3bd44a9cb49c7d1b6a4db642": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "9036a15be46e45799e8e3cc854fa88bb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "932c4d4ac22d402fa5f1564437dc5e5b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_35750f3fe46e40b99058d14ca7ed800b",
       "IPY_MODEL_f4e20a4a864b4c13adc38f22fccb80a0",
       "IPY_MODEL_3987a825b2584353a2be546f63911a0c"
      ],
      "layout": "IPY_MODEL_1bc226c51f954bbf98c13d2310940135"
     }
    },
    "9c8aa38595dc4b7e8ca640d03cc19694": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "9ce5c894beb448b1b5935907c38d9138": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "a02025fa486a4cab95eb79496bc3e7ea": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a0fd105ea5f046c18b8191d1ae5d89bc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a236061375e947dc9612c24adbe4ff0b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_6bdcd35840594df4ab40d4c64e660dba",
      "max": 116,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_f3384eb9244c4701ac8c1b16772132cf",
      "value": 116
     }
    },
    "a30e36f3bf9e4c3e85df561d3f1a81a1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_63dc8025965242e09dfb7fffb70c953a",
       "IPY_MODEL_4b8aa7e353f045dd8297de8c688b9d40",
       "IPY_MODEL_0cdbc7260ad346048a87a3f8ba22486e"
      ],
      "layout": "IPY_MODEL_59f57f5ddfd24b53a4f1621f1a6389bb"
     }
    },
    "a53e2fc70d75494da4d6106f6608875b": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ab01960dd6954af78441736c8669b241": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "abdbb22939fd41db9b0a3ee58c7ed13b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_7a839c9bc6ca43fdbf6d63cfe823af63",
      "placeholder": "​",
      "style": "IPY_MODEL_59d57ba6032a46d79628065ec01cff0f",
      "value": "special_tokens_map.json: 100%"
     }
    },
    "af86bd6b8bef45c99bc6d61af2c4f71f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b05eb88e352947e295a0ed84710ed5f0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b4717e6adc3e42648ffbf1e202a852ca": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b5a63beb6d2b4249871961e1a0a9b937": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b63ec0fd33c448e89cc03e6000abee00": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2153807b370049bd95c515e53f487f72",
      "max": 466247,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_291e805751f84bb3a29d25c8b7cb3204",
      "value": 466247
     }
    },
    "b73272664dfb4605b1d3a85928865304": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_879a060bd57146398f7c175704456532",
      "placeholder": "​",
      "style": "IPY_MODEL_64a3c90a4ba94283bbc717031722190c",
      "value": "config.json: 100%"
     }
    },
    "bd7036844925403fb3db2f1985fbf7f8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "bf34bae971834a2ca8ab1ac10e833986": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "bf354c1bbf524f97a07cd2a6c854e22c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c2058ed868ff4815a9ef7ea89e85f022": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c2d096efb1db4dde86fd42ff3a4df605": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c92270c929c24bcb847c6739686a799b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e5aa932cda80483d8328d9f942462e9e",
      "max": 112,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_e0e136927c3644c6bfb9d611817870c8",
      "value": 112
     }
    },
    "cd89e860240142e28da826f6f49ba240": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f412568cd7f741c5a2d4a18416981739",
      "placeholder": "​",
      "style": "IPY_MODEL_50b11ecfa6ee4d4d8d4fc97a31a4551b",
      "value": " 116/116 [00:00&lt;00:00, 3.09kB/s]"
     }
    },
    "d357f7f4bd3f4af7af517dd7a725d2f0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "dcbf618d0a85491f8254ead7a0c027f5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "e0e136927c3644c6bfb9d611817870c8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "e2c7cbcbf7cd4e0ca1352925b8c7bdea": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_30a058a01b434739adf517283900f1c8",
      "max": 53,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_9036a15be46e45799e8e3cc854fa88bb",
      "value": 53
     }
    },
    "e3f45f680ec84c4eb6ee402b187e38a8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "e48c72f5aa414987a2ce9409e9b5ad69": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e5aa932cda80483d8328d9f942462e9e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "efa59884be0e4fa6bb6ddf4d4885ab52": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_bd7036844925403fb3db2f1985fbf7f8",
      "max": 231508,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_33c5f2daa3a146e4ad66a753a4969e71",
      "value": 231508
     }
    },
    "f3384eb9244c4701ac8c1b16772132cf": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "f412568cd7f741c5a2d4a18416981739": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f4e20a4a864b4c13adc38f22fccb80a0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b05eb88e352947e295a0ed84710ed5f0",
      "max": 90868376,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_dcbf618d0a85491f8254ead7a0c027f5",
      "value": 90868376
     }
    },
    "f7d8f5d7761541ec95481008339f2420": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "f94f4f97ce194f77bdd6d4bdfea4edea": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "fb9ef2497d3b47a297501004b0176cde": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2377c5a7810c4e86849f2f1889cb0a5e",
      "placeholder": "​",
      "style": "IPY_MODEL_4a8445cda576479ab8663568b2e58fe0",
      "value": "tokenizer.json: 100%"
     }
    },
    "fbb3251aed0743e1be86088735d2ba33": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "fe74971a69824ace86c03a408afe7100": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_a0fd105ea5f046c18b8191d1ae5d89bc",
      "placeholder": "​",
      "style": "IPY_MODEL_2319ac540cdb44eba7715740e1f391b3",
      "value": " 190/190 [00:00&lt;00:00, 2.82kB/s]"
     }
    },
    "fec6c46a94314fff8f5fa6db76c722a2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "fffef1e3409c4e929c21060b07853091": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}