From ca0812802522d19b7e45935ab0f828d03d10dc2c Mon Sep 17 00:00:00 2001
From: abhiupes01 <39192050+abhiupes01@users.noreply.github.com>
Date: Sun, 30 Aug 2020 18:52:43 +0800
Subject: [PATCH] Add files via upload

---
 bayesian_neural_network.ipynb | 1179 +++++++++++++++++++++++++++++++++
 neural_network.ipynb          |  638 ++++++++++++++++++
 random_frst.ipynb             |  488 ++++++++++++++
 svm.ipynb                     |  292 ++++++++
 4 files changed, 2597 insertions(+)
 create mode 100644 bayesian_neural_network.ipynb
 create mode 100644 neural_network.ipynb
 create mode 100644 random_frst.ipynb
 create mode 100644 svm.ipynb

diff --git a/bayesian_neural_network.ipynb b/bayesian_neural_network.ipynb
new file mode 100644
index 0000000..75cbac5
--- /dev/null
+++ b/bayesian_neural_network.ipynb
@@ -0,0 +1,1179 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# importing the dependent libraries\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "%matplotlib inline\n",
+    "import tensorflow \n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.layers import Dense, Activation, Dropout\n",
+    "from sklearn.model_selection import GridSearchCV\n",
+    "from tensorflow.keras.wrappers.scikit_learn import KerasClassifier\n",
+    "from sklearn.preprocessing import MinMaxScaler\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.model_selection import learning_curve\n",
+    "from tensorflow.keras import optimizers\n",
+    "import seaborn as sns\n",
+    "from scipy import stats\n",
+    "from scipy.stats import skew, norm, probplot, boxcox\n",
+    "from scipy.special import boxcox1p\n",
+    "import scipy\n",
+    "import tensorflow as tf\n",
+    "from tensorflow import keras\n",
+    "\n",
+    "# Bayesian Optimization\n",
+    "import skopt\n",
+    "from skopt import gp_minimize, forest_minimize\n",
+    "from skopt.space import Real, Categorical, Integer\n",
+    "from skopt.plots import plot_convergence\n",
+    "from skopt.plots import plot_objective, plot_evaluations\n",
+    "from skopt.utils import use_named_args\n",
+    "#from skopt.plots import plot_histogram, plot_objective"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.read_csv('pressure_loss.csv')\n",
+    "data['diff_p'] = data['diff_p']*0.24884\n",
+    "data = data.drop(['annulus','test_fluid'],axis=1)\n",
+    "dataset = data.values\n",
+    "x = dataset[:,:6]\n",
+    "y = dataset[:,6:]\n",
+    "scale_1 = MinMaxScaler(feature_range=(0,1))\n",
+    "scale_2 = MinMaxScaler(feature_range=(0,1))\n",
+    "x_scale = scale_1.fit_transform(x)\n",
+    "y_scale = scale_2.fit_transform(y)\n",
+    "# splitting the data\n",
+    "x_train,x_val, y_train,y_val = train_test_split(x_scale,y_scale,test_size=0.3)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define learning rate for skopt\n",
+    "dim_learning_rate = Real(low=1e-6,high=1e-2,prior='log-uniform',name='learning_rate')\n",
+    "\n",
+    "# define num_neurons\n",
+    "dim_neurons = Integer(low=1,high=40,name='num_neurons')\n",
+    "\n",
+    "# define number of layers \n",
+    "dim_layers = Integer(low=1,high=40,name='num_layers')\n",
+    "\n",
+    "# define axtivation function\n",
+    "dim_activation = Categorical(categories=['relu','sigmoid','tanh'],name='act')\n",
+    "\n",
+    "# define the dropout \n",
+    "dim_dropout = Real(low=0.0001,high = 0.3,name='dropout')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define the total dimenison to be considered in the study\n",
+    "dimensions = [dim_neurons,\n",
+    "             dim_layers,\n",
+    "             dim_learning_rate,\n",
+    "             dim_activation,\n",
+    "             dim_dropout]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define the total dimenison to be considered in the study\n",
+    "dimensions = [dim_neurons,\n",
+    "             dim_layers,\n",
+    "             dim_learning_rate,\n",
+    "             dim_activation,\n",
+    "             dim_dropout]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Optimizing the ANN with Bayesian Optimization method\n",
+    "# generalized neural network\n",
+    "# Define the model for the bayesian optimization\n",
+    "# define a base case of ANN applicable for allthe cases and optimization\n",
+    "# importing the basic libraries\n",
+    "# Define the neural network\n",
+    "def create_model(num_neurons,num_layers,learning_rate,act,\n",
+    "                   dropout):\n",
+    "    \n",
+    "    \"\"\"\n",
+    "    Hyperparameters \n",
+    "    num_neurons:     Number of neurons in a layer\n",
+    "    num_layers:      Number of hidden layer in the ANN\n",
+    "    input_dim:       Shape of the input data into first layer\n",
+    "    output_dim:      Number of classes to be predicted\n",
+    "    learning_rate:   Learning rate of the optimizer\n",
+    "    activation:      Activation function of the layer\n",
+    "    optimization:    Optimization algorithms to be used\n",
+    "    dropout:         Define the rate of dropout \n",
+    "    \"\"\"\n",
+    "    \n",
+    "    #set random seed for reproducibility\n",
+    "    np.random.seed(42)\n",
+    "    #set_random_seed(42)\n",
+    "    \n",
+    "    # start the construction of keras Sequential Model\n",
+    "    model = Sequential()\n",
+    "    \n",
+    "    # Add an input layer \n",
+    "    model.add(Dense(num_neurons,activation='relu',input_dim=6))\n",
+    "    \n",
+    "    # Iterate for the additional layers\n",
+    "    for i in range(1,num_layers):\n",
+    "        model.add(Dense(num_neurons,activation=act))\n",
+    "        \n",
+    "    # Add the droppout layer\n",
+    "    model.add(Dropout(dropout))\n",
+    "        \n",
+    "    # add the last layer\n",
+    "    model.add(Dense(1,activation='linear'))\n",
+    "    \n",
+    "    # define the optimization function\n",
+    "    adam = optimizers.Adam(lr=learning_rate)\n",
+    "    \n",
+    "    # Compiling the model \n",
+    "    model.compile(adam,\n",
+    "                  loss='mean_squared_error',\n",
+    "                  metrics=['mse']\n",
+    "                 )\n",
+    "    return model\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# define the best accuracy of the model \n",
+    "best_accuracy = 0.0\n",
+    "path_best_model = 'show.best_model.keras'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Default parameters for the above described model\n",
+    "default_parameter = [7,3,0.001,'relu',0.001]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# coding the fitness function\n",
+    "\n",
+    "@use_named_args(dimensions=dimensions)\n",
+    "def fitness(num_neurons,num_layers,learning_rate,act,dropout):\n",
+    "    \n",
+    "    # print the hyperparameters\n",
+    "    print('Number of Neurons: ',num_neurons)\n",
+    "    print('Nuber of Layers: ',num_layers)\n",
+    "    print('Learning Rate: {0:.1e}'.format(learning_rate))\n",
+    "    print('Activation Function: ',act)\n",
+    "    print('Dropout: ',dropout)\n",
+    "    \n",
+    "    # create the neural network with these hyperparameters\n",
+    "    model = create_model(num_neurons=num_neurons\n",
+    "                         ,num_layers=num_layers\n",
+    "                         ,learning_rate=learning_rate\n",
+    "                         ,act=act\n",
+    "                         ,dropout=dropout)\n",
+    "    \n",
+    "    # use keras to train the model\n",
+    "    history = model.fit(x=x_train,\n",
+    "                       y=y_train,\n",
+    "                       epochs=5,\n",
+    "                       batch_size=16,\n",
+    "                       validation_split=0.20,\n",
+    "                       verbose=0)\n",
+    "    \n",
+    "    # get the classification accuracy on the validation dataset after last training \n",
+    "    accuracy = history.history['val_loss'][-1]\n",
+    "    \n",
+    "    # print the classification accuracy\n",
+    "    print()\n",
+    "    print('Loss: {0:.2%}'.format(accuracy))\n",
+    "    print()\n",
+    "    \n",
+    "    global best_accuracy\n",
+    "    if accuracy<best_accuracy:\n",
+    "        model.save(path_best_model)\n",
+    "        best_accurcay = accuracy\n",
+    "    del model\n",
+    "    return accuracy"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Number of Neurons:  7\n",
+      "Nuber of Layers:  3\n",
+      "Learning Rate: 1.0e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.001\n",
+      "\n",
+      "Loss: 1.70%\n",
+      "\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "0.0169874136519479"
+      ]
+     },
+     "execution_count": 54,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "fitness(x=default_parameter)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Number of Neurons:  7\n",
+      "Nuber of Layers:  3\n",
+      "Learning Rate: 1.0e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.001\n",
+      "\n",
+      "Loss: 3.97%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.65%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.66%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.57%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.64%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.56%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.64%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.57%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.57%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.58%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  38\n",
+      "Learning Rate: 8.5e-04\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.046889990268686725\n",
+      "\n",
+      "Loss: 4.55%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 4.17%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 2.3e-05\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.00699656687135628\n",
+      "\n",
+      "Loss: 2.55%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.0e-06\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 15.46%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.2e-04\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 9.96%\n",
+      "\n",
+      "Number of Neurons:  26\n",
+      "Nuber of Layers:  9\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0061249741482980935\n",
+      "\n",
+      "Loss: 0.27%\n",
+      "\n",
+      "Number of Neurons:  16\n",
+      "Nuber of Layers:  34\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.01686675200065847\n",
+      "\n",
+      "Loss: 4.55%\n",
+      "\n",
+      "Number of Neurons:  31\n",
+      "Nuber of Layers:  11\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  tanh\n",
+      "Dropout:  0.004108603435711055\n",
+      "\n",
+      "Loss: 0.80%\n",
+      "\n",
+      "Number of Neurons:  35\n",
+      "Nuber of Layers:  2\n",
+      "Learning Rate: 9.6e-05\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.06292353247001682\n",
+      "\n",
+      "Loss: 3.82%\n",
+      "\n",
+      "Number of Neurons:  2\n",
+      "Nuber of Layers:  35\n",
+      "Learning Rate: 3.1e-03\n",
+      "Activation Function:  tanh\n",
+      "Dropout:  0.08766867518341284\n",
+      "\n",
+      "Loss: 2.43%\n",
+      "\n",
+      "Number of Neurons:  39\n",
+      "Nuber of Layers:  10\n",
+      "Learning Rate: 2.6e-03\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.08262074630879455\n",
+      "\n",
+      "Loss: 4.97%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  40\n",
+      "Learning Rate: 1.7e-05\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.25680059606798905\n",
+      "\n",
+      "Loss: 37.95%\n",
+      "\n",
+      "Number of Neurons:  5\n",
+      "Nuber of Layers:  22\n",
+      "Learning Rate: 3.0e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.11263738798204785\n",
+      "\n",
+      "Loss: 0.23%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.2339227258010004\n",
+      "\n",
+      "Loss: 4.52%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 3.5e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 0.37%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  40\n",
+      "Learning Rate: 1.7e-05\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.24276521491446992\n",
+      "\n",
+      "Loss: 5.95%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  40\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 4.71%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 0.75%\n",
+      "\n",
+      "Number of Neurons:  27\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 3.1e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 0.91%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  8\n",
+      "Learning Rate: 1.0e-06\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.10407535445360852\n",
+      "\n",
+      "Loss: 6.87%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  30\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.1289080094107778\n",
+      "\n",
+      "Loss: 4.57%\n",
+      "\n",
+      "Number of Neurons:  17\n",
+      "Nuber of Layers:  33\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.11768834061933628\n",
+      "\n",
+      "Loss: 4.61%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 3.9e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.016492040322478342\n",
+      "\n",
+      "Loss: 4.05%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  37\n",
+      "Learning Rate: 6.4e-04\n",
+      "Activation Function:  tanh\n",
+      "Dropout:  0.21036054754616587\n",
+      "\n",
+      "Loss: 0.30%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  7\n",
+      "Learning Rate: 1.5e-03\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.20061487406562792\n",
+      "\n",
+      "Loss: 4.72%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  6\n",
+      "Learning Rate: 1.0e-06\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.08383419663425465\n",
+      "\n",
+      "Loss: 15.67%\n",
+      "\n",
+      "Number of Neurons:  17\n",
+      "Nuber of Layers:  35\n",
+      "Learning Rate: 1.5e-05\n",
+      "Activation Function:  tanh\n",
+      "Dropout:  0.2138389588619408\n",
+      "\n",
+      "Loss: 4.35%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  20\n",
+      "Learning Rate: 1.7e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 0.43%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  3\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.16272161916370845\n",
+      "\n",
+      "Loss: 0.28%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  8\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.2238006280407738\n",
+      "\n",
+      "Loss: 4.97%\n",
+      "\n",
+      "Number of Neurons:  1\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 4.58%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 5.1e-04\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.28767823578801566\n",
+      "\n",
+      "Loss: 1.73%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  13\n",
+      "Learning Rate: 1.6e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.13998523423646897\n",
+      "\n",
+      "Loss: 0.18%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.5e-05\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.16715286315316646\n",
+      "\n",
+      "Loss: 5.79%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  40\n",
+      "Learning Rate: 1.0e-06\n",
+      "Activation Function:  tanh\n",
+      "Dropout:  0.040704871824048486\n",
+      "\n",
+      "Loss: 3.55%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.0e-02\n",
+      "Activation Function:  sigmoid\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 0.62%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  23\n",
+      "Learning Rate: 1.6e-04\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.3\n",
+      "\n",
+      "Loss: 1.44%\n",
+      "\n",
+      "Number of Neurons:  13\n",
+      "Nuber of Layers:  1\n",
+      "Learning Rate: 1.0e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.2284714118692474\n",
+      "\n",
+      "Loss: 5.04%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  40\n",
+      "Learning Rate: 2.1e-05\n",
+      "Activation Function:  tanh\n",
+      "Dropout:  0.0001\n",
+      "\n",
+      "Loss: 1.77%\n",
+      "\n",
+      "Number of Neurons:  40\n",
+      "Nuber of Layers:  31\n",
+      "Learning Rate: 2.0e-03\n",
+      "Activation Function:  relu\n",
+      "Dropout:  0.3\n",
+      "\n",
+      "Loss: 4.58%\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "search_result = gp_minimize(func=fitness,\n",
+    "                           dimensions=dimensions,\n",
+    "                           acq_func='EI',\n",
+    "                           n_calls=50,\n",
+    "                           x0=default_parameter)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 56,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot the convergence of the fitness function\n",
+    "plot_convergence(search_result);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 57,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[40, 13, 0.0015985572878205064, 'relu', 0.13998523423646897]"
+      ]
+     },
+     "execution_count": 57,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# best hyper parameters\n",
+    "search_result.x"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 58,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.0018399823672416817"
+      ]
+     },
+     "execution_count": 58,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "space = search_result.space\n",
+    "search_result.fun"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[(0.0018399823672416817,\n",
+       "  [40, 13, 0.0015985572878205064, 'relu', 0.13998523423646897]),\n",
+       " (0.0023179996408536794,\n",
+       "  [5, 22, 0.0029506119371934827, 'relu', 0.11263738798204785]),\n",
+       " (0.002655855961158637, [26, 9, 0.01, 'relu', 0.0061249741482980935]),\n",
+       " (0.0027617777892073076, [40, 3, 0.01, 'sigmoid', 0.16272161916370845]),\n",
+       " (0.00300239134803561,\n",
+       "  [40, 37, 0.000643190894126107, 'tanh', 0.21036054754616587]),\n",
+       " (0.003661383575884493, [40, 1, 0.0035250454186137067, 'relu', 0.0001]),\n",
+       " (0.004258993177997374, [40, 20, 0.001731835155508845, 'relu', 0.0001]),\n",
+       " (0.006243824045811816, [40, 1, 0.01, 'sigmoid', 0.0001]),\n",
+       " (0.007516969589325856, [40, 1, 0.01, 'relu', 0.0001]),\n",
+       " (0.007980353558274705, [31, 11, 0.01, 'tanh', 0.004108603435711055]),\n",
+       " (0.009134037797815922, [27, 1, 0.003129110229952164, 'relu', 0.0001]),\n",
+       " (0.014377738829467475, [40, 23, 0.00015561205683012073, 'relu', 0.3]),\n",
+       " (0.017296143653472578,\n",
+       "  [40, 1, 0.0005145204388124426, 'relu', 0.28767823578801566]),\n",
+       " (0.01766880758209374, [40, 40, 2.0613333091720968e-05, 'tanh', 0.0001]),\n",
+       " (0.0243075679929003,\n",
+       "  [2, 35, 0.0030980825791273965, 'tanh', 0.08766867518341284]),\n",
+       " (0.025465207174420357,\n",
+       "  [1, 1, 2.25576177867712e-05, 'relu', 0.00699656687135628]),\n",
+       " (0.03547717617663342, [40, 40, 1e-06, 'tanh', 0.040704871824048486]),\n",
+       " (0.03821627317568448,\n",
+       "  [35, 2, 9.641380829949762e-05, 'relu', 0.06292353247001682]),\n",
+       " (0.03970049969910636, [7, 3, 0.001, 'relu', 0.001]),\n",
+       " (0.04045972154717746,\n",
+       "  [1, 1, 0.00392013982916096, 'relu', 0.016492040322478342]),\n",
+       " (0.04165366286133218, [1, 1, 0.01, 'relu', 0.0001]),\n",
+       " (0.043482556807126586,\n",
+       "  [17, 35, 1.5291635573964494e-05, 'tanh', 0.2138389588619408]),\n",
+       " (0.04523412950747595, [1, 1, 0.01, 'relu', 0.2339227258010004]),\n",
+       " (0.04548207258321638, [16, 34, 0.01, 'relu', 0.01686675200065847]),\n",
+       " (0.045483838781479775,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.04557745679332985,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.04567266653079217, [1, 30, 0.01, 'relu', 0.1289080094107778]),\n",
+       " (0.04571079958494254,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.045721271141307564,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.04573962523594616,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.04581180876340923, [1, 1, 0.01, 'sigmoid', 0.0001]),\n",
+       " (0.045843420447561685,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.04584512470920724, [40, 31, 0.0019750310005756558, 'relu', 0.3]),\n",
+       " (0.046082673741020555, [17, 33, 0.01, 'sigmoid', 0.11768834061933628]),\n",
+       " (0.04637229134599993,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.04639028178483952,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.04645158288868394,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.046562849976650375,\n",
+       "  [16, 38, 0.0008471801418819981, 'sigmoid', 0.046889990268686725]),\n",
+       " (0.047050225394447955, [1, 40, 0.01, 'relu', 0.0001]),\n",
+       " (0.047180764025121225,\n",
+       "  [1, 7, 0.0014814391177623426, 'sigmoid', 0.20061487406562792]),\n",
+       " (0.04966047783125573, [40, 8, 0.01, 'sigmoid', 0.2238006280407738]),\n",
+       " (0.04969101101453379,\n",
+       "  [39, 10, 0.002639592517907068, 'sigmoid', 0.08262074630879455]),\n",
+       " (0.050428153536333814,\n",
+       "  [13, 1, 0.0010377657252713952, 'relu', 0.2284714118692474]),\n",
+       " (0.05794348113062814,\n",
+       "  [40, 1, 1.4886070699064788e-05, 'relu', 0.16715286315316646]),\n",
+       " (0.059539444891251916,\n",
+       "  [40, 40, 1.6884424060135412e-05, 'relu', 0.24276521491446992]),\n",
+       " (0.06871783922565734, [40, 8, 1e-06, 'relu', 0.10407535445360852]),\n",
+       " (0.09955551404887297, [1, 1, 0.0001208532526390957, 'relu', 0.0001]),\n",
+       " (0.15461595297798397, [1, 1, 1e-06, 'relu', 0.0001]),\n",
+       " (0.15667050490229148, [1, 6, 1e-06, 'sigmoid', 0.08383419663425465]),\n",
+       " (0.3795376441140813,\n",
+       "  [40, 40, 1.7046985416263443e-05, 'sigmoid', 0.25680059606798905])]"
+      ]
+     },
+     "execution_count": 59,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sorted(zip(search_result.func_vals,search_result.x_iters))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0, 0.5, 'Number of Neurons')"
+      ]
+     },
+     "execution_count": 60,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 720x720 with 25 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig = plot_objective(result=search_result,\n",
+    "                       #dimension_name1='learning_rate',\n",
+    "                       #dimension_name2='num_neurons',\n",
+    "                       levels=50)\n",
+    "plt.xlabel('Learning Rate')\n",
+    "plt.ylabel('Number of Neurons')\n",
+    "#plt.savefig('lr_neurons.jpeg')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "test_model = create_model(35,16,0.006681727541638011,'relu',\n",
+    "                   0.030734643944295682)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class PrintDot(keras.callbacks.Callback):\n",
+    "    def on_epoch_end(self,epoch,logs):\n",
+    "        if epoch%100 ==0: print('')\n",
+    "        print('.',end='')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "...................................................................................................."
+     ]
+    }
+   ],
+   "source": [
+    "history_test = test_model.fit(x_train,y_train,batch_size=16, epochs=1000,validation_split=0.2,\n",
+    "                    verbose=0,callbacks=[PrintDot()])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "metadata": {},
+   "outputs": [
+    {
+     "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></th>\n",
+       "      <th>loss</th>\n",
+       "      <th>mse</th>\n",
+       "      <th>val_loss</th>\n",
+       "      <th>val_mse</th>\n",
+       "      <th>epoch</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>995</td>\n",
+       "      <td>0.000621</td>\n",
+       "      <td>0.000621</td>\n",
+       "      <td>0.001144</td>\n",
+       "      <td>0.001144</td>\n",
+       "      <td>995</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>996</td>\n",
+       "      <td>0.001008</td>\n",
+       "      <td>0.001008</td>\n",
+       "      <td>0.000703</td>\n",
+       "      <td>0.000703</td>\n",
+       "      <td>996</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>997</td>\n",
+       "      <td>0.000586</td>\n",
+       "      <td>0.000586</td>\n",
+       "      <td>0.000733</td>\n",
+       "      <td>0.000733</td>\n",
+       "      <td>997</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>998</td>\n",
+       "      <td>0.000618</td>\n",
+       "      <td>0.000618</td>\n",
+       "      <td>0.000682</td>\n",
+       "      <td>0.000682</td>\n",
+       "      <td>998</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>999</td>\n",
+       "      <td>0.001629</td>\n",
+       "      <td>0.001629</td>\n",
+       "      <td>0.000952</td>\n",
+       "      <td>0.000952</td>\n",
+       "      <td>999</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         loss       mse  val_loss   val_mse  epoch\n",
+       "995  0.000621  0.000621  0.001144  0.001144    995\n",
+       "996  0.001008  0.001008  0.000703  0.000703    996\n",
+       "997  0.000586  0.000586  0.000733  0.000733    997\n",
+       "998  0.000618  0.000618  0.000682  0.000682    998\n",
+       "999  0.001629  0.001629  0.000952  0.000952    999"
+      ]
+     },
+     "execution_count": 64,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "hist_test = pd.DataFrame(history_test.history)\n",
+    "hist_test['epoch'] = history_test.epoch\n",
+    "hist_test.tail()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 65,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "def plot_histry(history):\n",
+    "    hist = pd.DataFrame(history.history)\n",
+    "    hist['epoch'] = history.epoch\n",
+    "    \n",
+    "#     plt.figure()\n",
+    "#     plt.xlabel('Epoch')\n",
+    "#     plt.ylabel('Mean Absolute Error')\n",
+    "#     plt.plot(hist['epoch'],hist['mean_absolute_error'],label='Train Error')\n",
+    "#     plt.plot(hist['epoch'],hist['val_mean_absolute_error'],label='Val Error')\n",
+    "#     plt.legend()\n",
+    "    \n",
+    "#     plt.figure()\n",
+    "#     plt.xlabel('Epoch')\n",
+    "#     plt.ylabel('Mean Square Error')\n",
+    "#     plt.plot(hist['epoch'],hist['mean_squared_error'],label='Train Error')\n",
+    "#     plt.plot(hist['epoch'],hist['val_mean_squared_error'],label='Val Error')\n",
+    "#     plt.legend()\n",
+    "    \n",
+    "    plt.figure()\n",
+    "    plt.xlabel('Epoch')\n",
+    "    plt.ylabel('Loss')\n",
+    "    plt.plot(hist['epoch'],hist['loss'],label='Train Error')\n",
+    "    plt.plot(hist['epoch'],hist['val_loss'],label='Val Error')\n",
+    "    plt.legend()\n",
+    "\n",
+    "    \n",
+    "plot_histry(history_test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 67,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x1375f079288>"
+      ]
+     },
+     "execution_count": 67,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1152x432 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot trainning and test data \n",
+    "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(16,6))\n",
+    "#fig.suptitle('Bayesian Neural Network', fontsize=16)\n",
+    "fig.subplots_adjust(top=.85, wspace=0.2)\n",
+    "\n",
+    "# first track\n",
+    "ax[0].plot(scale_2.inverse_transform(test_model.predict(x_train)),scale_2.inverse_transform(y_train),'x',label='Data Point',ms=7)\n",
+    "ax[0].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].grid(True)\n",
+    "ax[0].set_title('(a) Training Data')\n",
+    "ax[0].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[0].tick_params(labelsize=15)\n",
+    "ax[0].legend()\n",
+    "\n",
+    "# second track\n",
+    "ax[1].plot(scale_2.inverse_transform(test_model.predict(x_val)),scale_2.inverse_transform(y_val),'x',label='Data Point',ms=7)\n",
+    "ax[1].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].grid(True)\n",
+    "ax[1].set_title('(b) Test Data')\n",
+    "ax[1].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[1].tick_params(labelsize=15)\n",
+    "ax[1].legend()\n",
+    "\n",
+    "#plt.savefig('Bayesian_ANN.jpeg',dpi=200)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.7.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/neural_network.ipynb b/neural_network.ipynb
new file mode 100644
index 0000000..a9fcd97
--- /dev/null
+++ b/neural_network.ipynb
@@ -0,0 +1,638 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# importing the dependent libraries\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "%matplotlib inline\n",
+    "import tensorflow \n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.layers import Dense, Activation, Dropout\n",
+    "from sklearn.model_selection import GridSearchCV\n",
+    "from tensorflow.keras.wrappers.scikit_learn import KerasClassifier\n",
+    "from sklearn.preprocessing import MinMaxScaler\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.model_selection import learning_curve\n",
+    "from tensorflow.keras import optimizers\n",
+    "import seaborn as sns\n",
+    "from scipy import stats\n",
+    "from scipy.stats import skew, norm, probplot, boxcox\n",
+    "from scipy.special import boxcox1p\n",
+    "import scipy\n",
+    "import tensorflow as tf\n",
+    "from tensorflow import keras"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.read_csv('pressure_loss.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "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></th>\n",
+       "      <th>flow_rate_gpm</th>\n",
+       "      <th>annulus</th>\n",
+       "      <th>ecc</th>\n",
+       "      <th>di/do</th>\n",
+       "      <th>test_fluid</th>\n",
+       "      <th>shear_pa</th>\n",
+       "      <th>k</th>\n",
+       "      <th>m</th>\n",
+       "      <th>diff_p</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>0</td>\n",
+       "      <td>18.50</td>\n",
+       "      <td>annulus_1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.27</td>\n",
+       "      <td>XCD-PAC3</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.99</td>\n",
+       "      <td>0.48</td>\n",
+       "      <td>37.62</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>17.67</td>\n",
+       "      <td>annulus_1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.27</td>\n",
+       "      <td>XCD-PAC3</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.99</td>\n",
+       "      <td>0.48</td>\n",
+       "      <td>36.97</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>2</td>\n",
+       "      <td>16.92</td>\n",
+       "      <td>annulus_1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.27</td>\n",
+       "      <td>XCD-PAC3</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.99</td>\n",
+       "      <td>0.48</td>\n",
+       "      <td>35.79</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>3</td>\n",
+       "      <td>16.09</td>\n",
+       "      <td>annulus_1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.27</td>\n",
+       "      <td>XCD-PAC3</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.99</td>\n",
+       "      <td>0.48</td>\n",
+       "      <td>34.97</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>4</td>\n",
+       "      <td>15.17</td>\n",
+       "      <td>annulus_1</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.27</td>\n",
+       "      <td>XCD-PAC3</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.99</td>\n",
+       "      <td>0.48</td>\n",
+       "      <td>34.18</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   flow_rate_gpm    annulus  ecc  di/do test_fluid  shear_pa     k     m  \\\n",
+       "0          18.50  annulus_1  1.0   0.27   XCD-PAC3       0.0  0.99  0.48   \n",
+       "1          17.67  annulus_1  1.0   0.27   XCD-PAC3       0.0  0.99  0.48   \n",
+       "2          16.92  annulus_1  1.0   0.27   XCD-PAC3       0.0  0.99  0.48   \n",
+       "3          16.09  annulus_1  1.0   0.27   XCD-PAC3       0.0  0.99  0.48   \n",
+       "4          15.17  annulus_1  1.0   0.27   XCD-PAC3       0.0  0.99  0.48   \n",
+       "\n",
+       "   diff_p  \n",
+       "0   37.62  \n",
+       "1   36.97  \n",
+       "2   35.79  \n",
+       "3   34.97  \n",
+       "4   34.18  "
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data['diff_p'] = data['diff_p']*0.24884"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = data.drop(['annulus','test_fluid'],axis=1)\n",
+    "dataset = data.values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "x = dataset[:,:6]\n",
+    "y = dataset[:,6:]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "scale_1 = MinMaxScaler(feature_range=(0,1))\n",
+    "scale_2 = MinMaxScaler(feature_range=(0,1))\n",
+    "x_scale = scale_1.fit_transform(x)\n",
+    "y_scale = scale_2.fit_transform(y)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# splitting the data\n",
+    "x_train,x_val, y_train,y_val = train_test_split(x_scale,y_scale,test_size=0.3)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(632, 6) (632, 1) (271, 6) (271, 1)\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(x_train.shape,y_train.shape,x_val.shape,y_val.shape)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create model in keras\n",
+    "model = Sequential([\n",
+    "    Dense(3,activation='relu',input_shape=(6,)),\n",
+    "    Dense(3,activation='relu'),\n",
+    "    Dropout(0.15),\n",
+    "    Dense(1,activation='linear'),\n",
+    "])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model.compile(optimizer='adam',\n",
+    "             loss='mean_squared_error',\n",
+    "              metrics=['mae','mape','mse']\n",
+    "             )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "....................................................................................................\n",
+      "...................................................................................................."
+     ]
+    }
+   ],
+   "source": [
+    "class PrintDot(keras.callbacks.Callback):\n",
+    "    def on_epoch_end(self,epoch,logs):\n",
+    "        if epoch%100 ==0: print('')\n",
+    "        print('.',end='')\n",
+    "history = model.fit(x_train,y_train,batch_size=16, epochs=1000,validation_split=0.2,\n",
+    "                    verbose=0,callbacks=[PrintDot()])\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [
+    {
+     "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></th>\n",
+       "      <th>loss</th>\n",
+       "      <th>mae</th>\n",
+       "      <th>mape</th>\n",
+       "      <th>mse</th>\n",
+       "      <th>val_loss</th>\n",
+       "      <th>val_mae</th>\n",
+       "      <th>val_mape</th>\n",
+       "      <th>val_mse</th>\n",
+       "      <th>epoch</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>995</td>\n",
+       "      <td>0.010564</td>\n",
+       "      <td>0.047308</td>\n",
+       "      <td>309.965698</td>\n",
+       "      <td>0.010564</td>\n",
+       "      <td>0.002076</td>\n",
+       "      <td>0.035507</td>\n",
+       "      <td>178.162796</td>\n",
+       "      <td>0.002076</td>\n",
+       "      <td>995</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>996</td>\n",
+       "      <td>0.008105</td>\n",
+       "      <td>0.042978</td>\n",
+       "      <td>324.355774</td>\n",
+       "      <td>0.008105</td>\n",
+       "      <td>0.002811</td>\n",
+       "      <td>0.037640</td>\n",
+       "      <td>160.652496</td>\n",
+       "      <td>0.002811</td>\n",
+       "      <td>996</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>997</td>\n",
+       "      <td>0.006266</td>\n",
+       "      <td>0.038945</td>\n",
+       "      <td>277.552246</td>\n",
+       "      <td>0.006266</td>\n",
+       "      <td>0.001840</td>\n",
+       "      <td>0.031400</td>\n",
+       "      <td>145.655746</td>\n",
+       "      <td>0.001840</td>\n",
+       "      <td>997</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>998</td>\n",
+       "      <td>0.009604</td>\n",
+       "      <td>0.042562</td>\n",
+       "      <td>283.434296</td>\n",
+       "      <td>0.009604</td>\n",
+       "      <td>0.002045</td>\n",
+       "      <td>0.033032</td>\n",
+       "      <td>151.482697</td>\n",
+       "      <td>0.002045</td>\n",
+       "      <td>998</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>999</td>\n",
+       "      <td>0.010761</td>\n",
+       "      <td>0.045774</td>\n",
+       "      <td>295.765106</td>\n",
+       "      <td>0.010761</td>\n",
+       "      <td>0.002703</td>\n",
+       "      <td>0.037256</td>\n",
+       "      <td>165.361694</td>\n",
+       "      <td>0.002703</td>\n",
+       "      <td>999</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         loss       mae        mape       mse  val_loss   val_mae    val_mape  \\\n",
+       "995  0.010564  0.047308  309.965698  0.010564  0.002076  0.035507  178.162796   \n",
+       "996  0.008105  0.042978  324.355774  0.008105  0.002811  0.037640  160.652496   \n",
+       "997  0.006266  0.038945  277.552246  0.006266  0.001840  0.031400  145.655746   \n",
+       "998  0.009604  0.042562  283.434296  0.009604  0.002045  0.033032  151.482697   \n",
+       "999  0.010761  0.045774  295.765106  0.010761  0.002703  0.037256  165.361694   \n",
+       "\n",
+       "      val_mse  epoch  \n",
+       "995  0.002076    995  \n",
+       "996  0.002811    996  \n",
+       "997  0.001840    997  \n",
+       "998  0.002045    998  \n",
+       "999  0.002703    999  "
+      ]
+     },
+     "execution_count": 34,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "hist = pd.DataFrame(history.history)\n",
+    "hist['epoch'] = history.epoch\n",
+    "hist.tail()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "def plot_histry(history):\n",
+    "    hist = pd.DataFrame(history.history)\n",
+    "    hist['epoch'] = history.epoch\n",
+    "    \n",
+    "    plt.figure()\n",
+    "    plt.xlabel('Epoch')\n",
+    "    plt.ylabel('Mean Absolute Error')\n",
+    "    plt.plot(hist['epoch'],hist['mae'],label='Train Error')\n",
+    "    plt.plot(hist['epoch'],hist['val_mae'],label='Val Error')\n",
+    "    plt.legend()\n",
+    "    \n",
+    "    plt.figure()\n",
+    "    plt.xlabel('Epoch')\n",
+    "    plt.ylabel('Mean Square Error')\n",
+    "    plt.plot(hist['epoch'],hist['mse'],label='Train Error')\n",
+    "    plt.plot(hist['epoch'],hist['val_mse'],label='Val Error')\n",
+    "    plt.legend()\n",
+    "    \n",
+    "    plt.figure()\n",
+    "    plt.xlabel('Epoch')\n",
+    "    plt.ylabel('Loss')\n",
+    "    plt.plot(hist['epoch'],hist['loss'],label='Train Error')\n",
+    "    plt.plot(hist['epoch'],hist['val_loss'],label='Val Error')\n",
+    "    plt.legend()\n",
+    "\n",
+    "    \n",
+    "plot_histry(history)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The mean squared error (MSE) for the test data set is: [0.003132138454319933, 0.038844895, 276.6478, 0.0031321384]\n"
+     ]
+    }
+   ],
+   "source": [
+    "test_error_rate = model.evaluate(x_train, y_train, verbose=0)\n",
+    "print(\"The mean squared error (MSE) for the test data set is: {}\".format(test_error_rate))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "271/271 [==============================] - 0s 37us/sample - loss: 0.0034 - mae: 0.0392 - mape: 179446.0156 - mse: 0.0034\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "[0.003409094803509512, 0.039213467, 179446.02, 0.0034090946]"
+      ]
+     },
+     "execution_count": 37,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model.evaluate(x_val,y_val)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "predict = model.predict(x_val)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y_pred = scale_2.inverse_transform(predict)\n",
+    "y_real = scale_2.inverse_transform(y_val)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1152x432 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot trainning and test data \n",
+    "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(16,6))\n",
+    "#fig.suptitle('Artificial Neural Network', fontsize=16)\n",
+    "fig.subplots_adjust(top=.85, wspace=0.2)\n",
+    "\n",
+    "\n",
+    "\n",
+    "# first track\n",
+    "ax[0].plot(scale_2.inverse_transform(model.predict(x_train)),scale_2.inverse_transform(y_train),'x',label='Data Point',ms=7)\n",
+    "ax[0].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].grid(True)\n",
+    "ax[0].set_title('(a) Training Data')\n",
+    "ax[0].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[0].legend()\n",
+    "ax[0].tick_params(labelsize=15)\n",
+    "\n",
+    "# second track\n",
+    "ax[1].plot(y_pred,y_real,'x',label='Data Point',ms=7)\n",
+    "ax[1].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].grid(True)\n",
+    "ax[1].set_title('(b) Test Data')\n",
+    "ax[1].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[1].legend()\n",
+    "ax[1].tick_params(labelsize=12)\n",
+    "#plt.rcParams.update({'font.size':10})\n",
+    "#plt.savefig('ANN.jpeg',dpi=200)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.7.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/random_frst.ipynb b/random_frst.ipynb
new file mode 100644
index 0000000..5b2a451
--- /dev/null
+++ b/random_frst.ipynb
@@ -0,0 +1,488 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "%matplotlib inline\n",
+    "import pylab\n",
+    "from sklearn import metrics\n",
+    "import seaborn as sns\n",
+    "from scipy import stats\n",
+    "from scipy.stats import skew, norm, probplot, boxcox\n",
+    "from scipy.special import boxcox1p\n",
+    "import scipy\n",
+    "from sklearn.metrics import r2_score\n",
+    "# plt.style.use('fivethirtyeight')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = pd.read_csv('pressure_loss.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data['diff_p'] = data['diff_p']*0.24884"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.ensemble import RandomForestRegressor"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "target = data['diff_p']\n",
+    "data_train = data.drop(['annulus','test_fluid','diff_p'],axis=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "feature_list = list(data_train.columns)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# convert data frame to np.arrays for processing\n",
+    "target = np.array(target)\n",
+    "data_train = np.array(data_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# splitting the data into train, test \n",
+    "x_train, x_test, y_train, y_test = train_test_split(data_train, target, random_state=42, test_size=0.3)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
+       "                      max_features='auto', max_leaf_nodes=None,\n",
+       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
+       "                      min_samples_leaf=1, min_samples_split=2,\n",
+       "                      min_weight_fraction_leaf=0.0, n_estimators=5, n_jobs=None,\n",
+       "                      oob_score=False, random_state=None, verbose=0,\n",
+       "                      warm_start=False)"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "rf = RandomForestRegressor(n_estimators=5)\n",
+    "rf.fit(x_train,y_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "predict = rf.predict(x_test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.9787345567896495"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "r2_score(predict,y_test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Variable: di/do                Importance: 0.58\n",
+      "Variable: flow_rate_gpm        Importance: 0.37\n",
+      "Variable: k                    Importance: 0.04\n",
+      "Variable: ecc                  Importance: 0.0\n",
+      "Variable: shear_pa             Importance: 0.0\n",
+      "Variable: m                    Importance: 0.0\n"
+     ]
+    }
+   ],
+   "source": [
+    "# feature importance\n",
+    "importances = list(rf.feature_importances_)\n",
+    "\n",
+    "#list of tuples with variables and importances\n",
+    "feature_importance = [(feature,round(importance,2)) for feature,importance in zip(feature_list,importances)]\n",
+    "\n",
+    "feature_importance = sorted(feature_importance,key=lambda x: x[1],reverse=True)\n",
+    "\n",
+    "[print('Variable: {:20} Importance: {}'.format(*pair)) for pair in feature_importance];"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "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></th>\n",
+       "      <th>0</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>Q (gpm)</td>\n",
+       "      <td>0.370326</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>e</td>\n",
+       "      <td>0.002366</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>Di/Do</td>\n",
+       "      <td>0.583853</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>τ_y</td>\n",
+       "      <td>0.002451</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>K</td>\n",
+       "      <td>0.036935</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>n</td>\n",
+       "      <td>0.004070</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                0\n",
+       "Q (gpm)  0.370326\n",
+       "e        0.002366\n",
+       "Di/Do    0.583853\n",
+       "τ_y      0.002451\n",
+       "K        0.036935\n",
+       "n        0.004070"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# feature = pd.DataFrame(data=np.round(list(rf.feature_importances_),2),columns=feature_list)\n",
+    "imp_feature=pd.DataFrame(list(rf.feature_importances_))\n",
+    "#imp_feature = imp_feature.T\n",
+    "imp_feature.index = ['Q (gpm)','e','Di/Do','τ_y','K','n']\n",
+    "imp_feature"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "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></th>\n",
+       "      <th>0</th>\n",
+       "      <th>percentage</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>Q (gpm)</td>\n",
+       "      <td>0.370326</td>\n",
+       "      <td>37.032611</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>e</td>\n",
+       "      <td>0.002366</td>\n",
+       "      <td>0.236550</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>Di/Do</td>\n",
+       "      <td>0.583853</td>\n",
+       "      <td>58.385273</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>τ_y</td>\n",
+       "      <td>0.002451</td>\n",
+       "      <td>0.245052</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>K</td>\n",
+       "      <td>0.036935</td>\n",
+       "      <td>3.693533</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>n</td>\n",
+       "      <td>0.004070</td>\n",
+       "      <td>0.406980</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                0  percentage\n",
+       "Q (gpm)  0.370326   37.032611\n",
+       "e        0.002366    0.236550\n",
+       "Di/Do    0.583853   58.385273\n",
+       "τ_y      0.002451    0.245052\n",
+       "K        0.036935    3.693533\n",
+       "n        0.004070    0.406980"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "imp_feature['percentage']=imp_feature[0]*100\n",
+    "imp_feature"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['Q (gpm)', 'e', 'Di/Do', 'τ_y', 'K', 'n']"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "labels = imp_feature.index.to_list()\n",
+    "labels "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.figure(figsize=(15,8))\n",
+    "imp_feature['percentage'].sort_values(ascending=False).plot(kind='bar')\n",
+    "plt.xlabel('Experimental Parameters',fontsize=18)\n",
+    "#plt.xticks(imp_feature['percentage'],imp_feature.index, rotation='vertical')\n",
+    "plt.ylabel('Importance Percentage',fontsize=18)\n",
+    "plt.title('Feature Importance',fontsize=20)\n",
+    "plt.xticks(rotation=0)\n",
+    "plt.tick_params(labelsize=20)\n",
+    "plt.savefig('feature_importance.jpeg',dpi=200)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x22e3fd25b88>"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1152x432 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot trainning and test data \n",
+    "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(16,6))\n",
+    "#fig.suptitle('Random Forest', fontsize=16)\n",
+    "fig.subplots_adjust(top=.85, wspace=0.2)\n",
+    "\n",
+    "# first track\n",
+    "ax[0].plot((rf.predict(x_train)),(y_train),'x',label='Data Point',ms=7)\n",
+    "ax[0].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].grid(True)\n",
+    "ax[0].set_title('(a) Training Data')\n",
+    "ax[0].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[0].tick_params(labelsize=15)\n",
+    "ax[0].legend()\n",
+    "\n",
+    "# second track\n",
+    "ax[1].plot(y_test,predict,'x',label='Data Point',ms=7)\n",
+    "ax[1].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].grid(True)\n",
+    "ax[1].set_title('(b) Test Data')\n",
+    "ax[1].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[1].tick_params(labelsize=15)\n",
+    "ax[1].legend()\n",
+    "\n",
+    "#plt.savefig('RF.jpeg',dpi=200)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.7.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/svm.ipynb b/svm.ipynb
new file mode 100644
index 0000000..dd2cbfe
--- /dev/null
+++ b/svm.ipynb
@@ -0,0 +1,292 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "%matplotlib inline\n",
+    "from sklearn import metrics\n",
+    "from sklearn.preprocessing import MinMaxScaler"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "press_data = pd.read_csv('pressure_loss.csv')\n",
+    "press_data['diff_p'] = press_data['diff_p']*0.24884\n",
+    "press_data.drop(['annulus','test_fluid'],axis=1,inplace=True)\n",
+    "col = [col for col in press_data.columns if col not in['diff_p']]\n",
+    "data = press_data[col]\n",
+    "target = press_data['diff_p']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "scale = MinMaxScaler()\n",
+    "scale_1 = MinMaxScaler()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data = data.values\n",
+    "target = target.values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "target.shape = (903,1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data_scale = scale.fit_transform(data)\n",
+    "target_scale = scale_1.fit_transform(target)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# split the data\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "data_train,data_test,target_train,target_test = train_test_split(data_scale,target_scale,test_size=0.3,random_state=42)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "((632, 6), (271, 6), (632, 1), (271, 1))"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# shape of the data set after partition\n",
+    "data_train.shape,data_test.shape,target_train.shape,target_test.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# import SVM library\n",
+    "from sklearn.svm import SVR\n",
+    "# metric \n",
+    "from sklearn.metrics import accuracy_score"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "SVR(C=1.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1, gamma='auto',\n",
+       "    kernel='rbf', max_iter=-1, shrinking=True, tol=0.001, verbose=False)"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# simple svr without any constraints\n",
+    "regressor = SVR(gamma='auto',)\n",
+    "regressor.fit(data_train,target_train.ravel())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pred = regressor.predict(data_test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.8390948567825165\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(regressor.score(data_test,target_test.ravel()))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "R2 Score: 0.9205761837926465\n"
+     ]
+    }
+   ],
+   "source": [
+    "# fine tune the SVRmodel\n",
+    "from sklearn.metrics import r2_score\n",
+    "regressor = SVR(gamma=1,kernel='rbf',degree=3,C=1)\n",
+    "pred = regressor.fit(data_train,target_train.ravel()).predict(data_test)\n",
+    "print('R2 Score:',r2_score(target_test,pred))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y_pred = scale_1.inverse_transform(pred.reshape(271,1))\n",
+    "y_real = scale_1.inverse_transform(target_test)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y_train_test = regressor.predict(data_train)\n",
+    "y_train_test = y_train_test.reshape(-1,632)\n",
+    "y_train_test = scale_1.inverse_transform(y_train_test)\n",
+    "y_target_test = scale_1.inverse_transform(target_train)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x2077e5af648>"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1152x432 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# plot trainning and test data \n",
+    "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(16,6))\n",
+    "#fig.suptitle('Support Vector Machine', fontsize=16)\n",
+    "fig.subplots_adjust(top=.85, wspace=0.2)\n",
+    "\n",
+    "# first track\n",
+    "ax[0].plot(y_target_test,y_train_test.reshape(632,-1),'x',label='Data Point',ms=7)\n",
+    "ax[0].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[0].grid(True)\n",
+    "ax[0].set_title('(a) Training Data')\n",
+    "ax[0].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[0].tick_params(labelsize=15)\n",
+    "ax[0].legend()\n",
+    "\n",
+    "# second track\n",
+    "ax[1].plot(y_real,y_pred,'x',label='Data Point',ms=7)\n",
+    "ax[1].set_ylabel('Predicted Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].set_xlabel('Observed Pressure Loss (kPa)',fontsize=15)\n",
+    "ax[1].grid(True)\n",
+    "ax[1].set_title('(b) Test Data')\n",
+    "ax[1].plot([0,400],[0,400],'--',label='Best Fit')\n",
+    "ax[1].tick_params(labelsize=15)\n",
+    "ax[1].legend()\n",
+    "\n",
+    "#plt.savefig('SVM.jpeg',dpi=200)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.7.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}