ULMFiT_fastai_Text_Classification.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ULMFiT_fastai_Text_Classification",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/navneetkrc/Colab_fastai/blob/master/ULMFiT_fastai_Text_Classification.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "rsMgf8LLsoOq",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Let's install PyTorch and fastai libraries first. You have to repeat this step every time you restart this notebook in colab."
      ]
    },
    {
      "metadata": {
        "id": "xlh_eIFNHS55",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "[Analytics Vidhya](https://www.analyticsvidhya.com/blog/2018/11/tutorial-text-classification-ulmfit-fastai-library/) Tutorial on which this Notebook is based on."
      ]
    },
    {
      "metadata": {
        "id": "9L-mYNWDBB4g",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "!pip install torch_nightly -f https://download.pytorch.org/whl/nightly/cu92/torch_nightly.html\n",
        "!pip install fastai"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "SA87AbrrBigr",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# import libraries\n",
        "import fastai\n",
        "from fastai import *\n",
        "from fastai.text import * \n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from functools import partial\n",
        "import io\n",
        "import os"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "r9SHCfiABlcF",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#import dataset\n",
        "from sklearn.datasets import fetch_20newsgroups\n",
        "dataset = fetch_20newsgroups(shuffle=True, random_state=1, remove=('headers', 'footers', 'quotes'))\n",
        "documents = dataset.data"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Xvs8MygLEKri",
        "colab_type": "code",
        "outputId": "df053b83-ca18-421e-dddf-ac046e6e6b2c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        }
      },
      "cell_type": "code",
      "source": [
        "dataset.target_names"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['alt.atheism',\n",
              " 'comp.graphics',\n",
              " 'comp.os.ms-windows.misc',\n",
              " 'comp.sys.ibm.pc.hardware',\n",
              " 'comp.sys.mac.hardware',\n",
              " 'comp.windows.x',\n",
              " 'misc.forsale',\n",
              " 'rec.autos',\n",
              " 'rec.motorcycles',\n",
              " 'rec.sport.baseball',\n",
              " 'rec.sport.hockey',\n",
              " 'sci.crypt',\n",
              " 'sci.electronics',\n",
              " 'sci.med',\n",
              " 'sci.space',\n",
              " 'soc.religion.christian',\n",
              " 'talk.politics.guns',\n",
              " 'talk.politics.mideast',\n",
              " 'talk.politics.misc',\n",
              " 'talk.religion.misc']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "metadata": {
        "id": "A3zbA9rfCfRh",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# create a dataframe\n",
        "df = pd.DataFrame({'label':dataset.target,\n",
        "                   'text':dataset.data})"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "lifPA3p6C853",
        "colab_type": "code",
        "outputId": "04e050fa-1a92-481e-bd34-d5bfd90580c4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "df.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(11314, 2)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "metadata": {
        "id": "1P5OelaiCmna",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "df = df[df['label'].isin([1,10])]\n",
        "df = df.reset_index(drop = True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "6zbrKEr7C419",
        "colab_type": "code",
        "outputId": "2d87cc65-a405-4b1c-e956-391a9dd61f7a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "df['label'].value_counts()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10    600\n",
              "1     584\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "metadata": {
        "id": "Z-jlCTiskI3X",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# split data into training and validation set\n",
        "df_trn, df_val = train_test_split(df, stratify = df['label'], test_size = 0.4, random_state = 12)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "EnpCz6vdkMY9",
        "colab_type": "code",
        "outputId": "85b4a291-b863-4e8e-cbf6-6e4a8a9d645f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "df_trn.shape, df_val.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((710, 2), (474, 2))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "metadata": {
        "id": "_SXRszhllc-E",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Preprocessing\n",
        "\n",
        "Not needed as we do not need tokens as we are going to use the sentence embeddings and the complete sentence matters."
      ]
    },
    {
      "metadata": {
        "id": "xvvFWWM4ljoH",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Data Preparation"
      ]
    },
    {
      "metadata": {
        "id": "7tNP_aMTEOKe",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Language model data\n",
        "data_lm = TextLMDataBunch.from_df(train_df = df_trn, valid_df = df_val, path = \"\")\n",
        "\n",
        "# Classifier model data\n",
        "data_clas = TextClasDataBunch.from_df(path = \"\", train_df = df_trn, valid_df = df_val, vocab=data_lm.train_ds.vocab, bs=32)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "8Ua8pmHEEP5N",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "learn = language_model_learner(data_lm, arch=AWD_LSTM, drop_mult=0.3)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "w3eJMLWDKb9u",
        "colab_type": "code",
        "outputId": "4bac07bb-1cf5-4606-ed4a-ee99ae161f49",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 378
        }
      },
      "cell_type": "code",
      "source": [
        "learn.lr_find()\n",
        "learn.recorder.plot()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              ""
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFYCAYAAAB6RnQAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nXOiffSIm5m0AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "YE3likAPETlz",
        "colab_type": "code",
        "outputId": "ff40c925-2ec6-4477-db4d-c5068866d42a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 165
        }
      },
      "cell_type": "code",
      "source": [
        "# train the learner object\n",
        "learn.fit_one_cycle(5, 1e-2)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "Total time: 01:07 <p><table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>1</td>\n",
              "      <td>4.981016</td>\n",
              "      <td>4.111104</td>\n",
              "      <td>0.266584</td>\n",
              "      <td>00:22</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2</td>\n",
              "      <td>4.193675</td>\n",
              "      <td>3.811648</td>\n",
              "      <td>0.290856</td>\n",
              "      <td>00:22</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3</td>\n",
              "      <td>3.734485</td>\n",
              "      <td>3.760023</td>\n",
              "      <td>0.296677</td>\n",
              "      <td>00:22</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "x7fO0cuVET8W",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "learn.save_encoder('ft_enc')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "uSkXI7HJET_r",
        "colab_type": "code",
        "outputId": "9db88a08-58b4-4dcd-d8d0-19cf73271331",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "my_fbeta = FBeta(beta=1,average='macro')\n",
        "learn.metrics = [accuracy, my_fbeta]\n",
        "learn.metrics"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<function fastai.metrics.accuracy>,\n",
              " FBeta(average='macro', pos_label=1, eps=1e-09, beta=1)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "metadata": {
        "id": "VsyHbNhc8_wn",
        "colab_type": "code",
        "outputId": "ff659d8b-1f90-4125-f89e-baadfa5e058f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "learn = text_classifier_learner(data_clas,arch=AWD_LSTM, drop_mult=0.7,metrics = [accuracy, my_fbeta])\n",
        "learn.load_encoder('ft_enc')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<function fastai.metrics.accuracy>,\n",
              " FBeta(average='macro', pos_label=1, eps=1e-09, beta=1)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "metadata": {
        "id": "1VaunwDAMmLV",
        "colab_type": "code",
        "outputId": "4b90724c-c6f7-4001-ec09-4baf132baa2c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 378
        }
      },
      "cell_type": "code",
      "source": [
        "learn.lr_find()\n",
        "learn.recorder.plot()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              ""
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfsAAAFYCAYAAABUA1WSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nL4QQQjg4afZCCCGEg/sfPkMtU/YpQEUAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "s7GTTw1GMl2F",
        "colab_type": "code",
        "outputId": "c1cef610-e40a-4b52-85a5-b2b4fc6f3094",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 165
        }
      },
      "cell_type": "code",
      "source": [
        "learn.fit_one_cycle(3, 1e-2)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "Total time: 03:22 <p><table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: left;\">\n",
              "      <th>epoch</th>\n",
              "      <th>train_loss</th>\n",
              "      <th>valid_loss</th>\n",
              "      <th>accuracy</th>\n",
              "      <th>f_beta</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>1</td>\n",
              "      <td>0.554966</td>\n",
              "      <td>0.349806</td>\n",
              "      <td>0.835443</td>\n",
              "      <td>0.831557</td>\n",
              "      <td>01:07</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2</td>\n",
              "      <td>0.508379</td>\n",
              "      <td>0.192840</td>\n",
              "      <td>0.945148</td>\n",
              "      <td>0.944956</td>\n",
              "      <td>01:07</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3</td>\n",
              "      <td>0.482716</td>\n",
              "      <td>0.165374</td>\n",
              "      <td>0.951477</td>\n",
              "      <td>0.951341</td>\n",
              "      <td>01:08</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "ggtcc_4ol9DH",
        "colab_type": "code",
        "outputId": "2e45902b-a71c-4988-a337-5268576931bb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 142
        }
      },
      "cell_type": "code",
      "source": [
        "# get predictions\n",
        "preds, targets = learn.get_preds()\n",
        "\n",
        "predictions = np.argmax(preds, axis = 1)\n",
        "pd.crosstab(predictions, targets)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th>col_0</th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>row_0</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>213</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>21</td>\n",
              "      <td>238</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "col_0    0    1\n",
              "row_0          \n",
              "0      213    2\n",
              "1       21  238"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "metadata": {
        "id": "LuC3tGFfmKOe",
        "colab_type": "code",
        "outputId": "9dce6c8e-1d38-4c48-c9b3-314b80400868",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "print(\"Process completed \")\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Process completed \n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}