deep learning intro with keras - linear classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import Input, Dense\n",
    "import matplotlib.image as img\n",
    "from scipy.cluster.vq import kmeans, vq, whiten\n",
    "import seaborn as sns, pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "from PIL import Image\n",
    "from keras.models import Sequential\n",
    "import tensorflow as tf\n",
    "from keras.models import Model\n",
    "from keras.utils import plot_model\n",
    "from sklearn.model_selection import train_test_split\n",
    "import numpy as np\n",
    "\n",
    "from numpy import array\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from keras.layers.embeddings import Embedding\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from keras.layers import Dense\n",
    "from keras.layers import Flatten\n",
    "\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from sklearn.metrics import classification_report,confusion_matrix\n",
    "from keras.optimizers import Adam\n",
    "from keras.models import Model\n",
    "from scipy.special import expit as sigmoid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv('wine.data.csv', usecols=['Cultivator', 'Alcohol', 'Malic_Acid', 'Ash', 'Alcalinity_of_Ash',\n",
    "       'Magnesium', 'Total_Phenols', 'Flavanoids', 'Nonflavanoid_phenols',\n",
    "       'Proanthocyanins', 'Color_Intensity', 'Hue', 'OD280', 'Proline'])\n",
    "\n",
    "#rint(df.keys())\n",
    "#print(df.head())\n",
    "\n",
    "X = df.drop('Cultivator',axis=1)\n",
    "y = df['Cultivator']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y,test_size=0.3, random_state=42)\n",
    "\n",
    "\n",
    "scaler = StandardScaler()\n",
    "\n",
    "scaler.fit(X_train)\n",
    "\n",
    "StandardScaler(copy=True, with_mean=True,with_std=True)\n",
    "X_train=scaler.transform(X_train)\n",
    "X_test=scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_19\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_39 (Dense)             (None, 13)                182       \n",
      "_________________________________________________________________\n",
      "dense_40 (Dense)             (None, 50)                700       \n",
      "_________________________________________________________________\n",
      "dense_41 (Dense)             (None, 1)                 51        \n",
      "=================================================================\n",
      "Total params: 933\n",
      "Trainable params: 933\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "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": [
    "model= Sequential()\n",
    "\n",
    "model.add(Dense(13, input_shape=(13,),\n",
    "#activation='sigmoid'\n",
    "activation='relu'\n",
    "))\n",
    "\n",
    "model.add(Dense(50, activation='relu'))\n",
    "model.add(Dense(1))\n",
    "\n",
    "#model.compile(optimizer=Adam(0.01),loss='binary_crossentropy')\n",
    "model.compile(optimizer=Adam(0.01),loss='mae')\n",
    "\n",
    "model.summary()\n",
    "\n",
    "plot_model(model, to_file='model.png')\n",
    "img=plt.imread('model.png')\n",
    "plt.imshow(img)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1\n",
      "124/124 [==============================] - 0s 56us/step - loss: 0.0276\n"
     ]
    }
   ],
   "source": [
    "#print(X_train[0])\n",
    "model.fit(X_train, y_train)\n",
    "\n",
    "predictions = model.predict(X_test)\n",
    "#cm=confusion_matrix(y_test,predictions)\n",
    "#print(cm)\n",
    "  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "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"
    }
   ],
   "source": [
    "y_pos = np.arange(len(predictions))\n",
    "\n",
    "prediction_result = predictions.flatten().round() \n",
    "\n",
    "plt.plot(predictions.round())\n",
    "#plt.barh(y_pos,list_test)\n",
    "plt.show()\n",
    "\n",
    "plt.bar(y_pos,prediction_result, color='green')\n",
    "plt.bar(y_pos,y_test,color='red',alpha=.6 )\n",
    "\n",
    "plt.show()\n",
    "#plt.xticks(np.arange(1, 1 + 2 * top_features), feature_names[top_coefficients], rotation=90, ha='right')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "54/54 [==============================] - 0s 259us/step\n",
      "0.08981275586066423\n"
     ]
    }
   ],
   "source": [
    "print(model.evaluate(X_test,y_test))\n",
    "#produces the mean square error of the system"
   ]
  },
  {
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}