{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "#para dividir los datos en train y test\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.metrics import classification_report,confusion_matrix\n",
    "\n",
    "\n",
    "\n",
    "train = pd.read_csv('data/fashion-mnist_train.csv')\n",
    "\n",
    "#por ahora no lo estamos usando\n",
    "test = pd.read_csv('data/fashion-mnist_test.csv')\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train.head()\n",
    "\n",
    "train.describe().transpose()\n",
    "\n",
    "X = train.drop('label',axis=1)\n",
    "\n",
    "y = train['label']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#divide los datos de entrenamiento y los datos de pruebas\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "#normalizacion de datos\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Fit only to the training data\n",
    "scaler.fit(X_train)\n",
    "\n",
    "StandardScaler(copy=True, with_mean=True, with_std=True)\n",
    "\n",
    "\n",
    "\n",
    "# Now apply the transformations to the data:\n",
    "X_train = scaler.transform(X_train)\n",
    "X_test = scaler.transform(X_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\n",
    "#no esta definida muy inteligentemente la cantidad de capas ni de neuronas\n",
    "#modelo de mi clasificador\n",
    "mlp = MLPClassifier(hidden_layer_sizes=(24,24,24,24,24),max_iter=500)\n",
    "\n",
    "#entrenamiento\n",
    "mlp.fit(X_train,y_train)\n",
    "\n",
    "predictions = mlp.predict(X_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1226    4   20   54    5    0  193    0   13    0]\n",
      " [  10 1498    0   19    3    1    4    0    3    0]\n",
      " [  17    5 1121   15  175    0  122    0    8    0]\n",
      " [  64   16   20 1311   46    1   38    0   12    0]\n",
      " [   5    6  112   50 1149    0  138    0   12    0]\n",
      " [   4    1    1    0    0 1462    4   33    5   18]\n",
      " [ 214    5  127   42   99    0 1029    0   13    1]\n",
      " [   0    0    0    0    0   58    0 1420    6   51]\n",
      " [  22    7    9    8   11   16   29    4 1390    1]\n",
      " [   0    0    0    0    0   20    0   56    1 1337]]\n"
     ]
    }
   ],
   "source": [
    "print(confusion_matrix(y_test,predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.78      0.81      0.80      1515\n",
      "          1       0.97      0.97      0.97      1538\n",
      "          2       0.80      0.77      0.78      1463\n",
      "          3       0.87      0.87      0.87      1508\n",
      "          4       0.77      0.78      0.78      1472\n",
      "          5       0.94      0.96      0.95      1528\n",
      "          6       0.66      0.67      0.67      1530\n",
      "          7       0.94      0.93      0.93      1535\n",
      "          8       0.95      0.93      0.94      1497\n",
      "          9       0.95      0.95      0.95      1414\n",
      "\n",
      "avg / total       0.86      0.86      0.86     15000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}