Test for Otto Group Problem.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "adb8f496-754a-4cf3-a359-459bc7f0db76",
   "metadata": {},
   "source": [
    "# ML Hackaton                  otto group multi-classification\n",
    "# Total number of nine classes, but about 93 features...\n",
    "# MLP/softmax function, using ReLU as activition function\n",
    "# logistic regression?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4808e2b7-22cb-44b4-8ebd-3db7c3f3f6e9",
   "metadata": {},
   "source": [
    "# How to determine useful features without training or using L1 regularization to do feature selection?\n",
    "# Correlation between features?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "0c06bb71-e63e-4383-8820-090dafb161b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import sklearn\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.datasets import make_classification\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.utils import resample\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "id": "ec4dcccc-c78c-4a5d-9348-25bd87e0d25d",
   "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": [
    "d = {\"Class_1\": 0, \"Class_2\": 1, \"Class_3\": 2, \"Class_4\": 3, \"Class_5\": 4, \"Class_6\": 5, \"Class_7\": 6, \"Class_8\": 7, \"Class_9\": 8,}\n",
    "df = pd.read_csv(\"Data for Otto Group/train.csv\")\n",
    "df[\"target\"] = df[\"target\"].map(d)\n",
    "df[\"target\"].value_counts().plot.bar()\n",
    "df = df.drop(\"id\", 1)\n",
    "X = df.drop(\"target\", 1)\n",
    "Y = df[\"target\"]\n",
    "# the graph below shows imbalanced labels in dataset, which may affects prediction of this model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "id": "2a25370d-da92-40f3-a05b-2e0e32a16110",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\n# resamplnig data to balance labels\\ndf_majority = df[df.target == 1]\\ndf_minority = df[df.target != 1]\\nl = [0, 2, 3, 4, 5, 6, 7, 8]\\n\\n\\nfor value in l:\\n    df_minority_resampled = resample(df_minority[df.target == value], replace=True, n_samples=16000, random_state=123)\\n    df_majority = pd.concat([df_minority_resampled, df_majority])\\n\\n#df_balanced = pd.concat([df_minority_resampled, df_majority])\\n\\ndf_majority[\"target\"].value_counts().plot.bar()\\n\\nX = df_majority.drop(\"target\", 1)\\nY = df_majority[\"target\"]\\n#df_upsample_balanced = pd.concat([df_minority_resampled, df_minority])\\n#df_upsample_balanced[\"target\"].value_counts().plot.bar()\\n'"
      ]
     },
     "execution_count": 223,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''   this doesn't work...\n",
    "# resamplnig data to balance labels\n",
    "df_majority = df[df.target == 1]\n",
    "df_minority = df[df.target != 1]\n",
    "l = [0, 2, 3, 4, 5, 6, 7, 8]\n",
    "\n",
    "\n",
    "for value in l:\n",
    "    df_minority_resampled = resample(df_minority[df.target == value], replace=True, n_samples=16000, random_state=123)\n",
    "    df_majority = pd.concat([df_minority_resampled, df_majority])\n",
    "\n",
    "#df_balanced = pd.concat([df_minority_resampled, df_majority])\n",
    "\n",
    "df_majority[\"target\"].value_counts().plot.bar()\n",
    "\n",
    "X = df_majority.drop(\"target\", 1)\n",
    "Y = df_majority[\"target\"]\n",
    "#df_upsample_balanced = pd.concat([df_minority_resampled, df_minority])\n",
    "#df_upsample_balanced[\"target\"].value_counts().plot.bar()\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "id": "f7d04479-2d6b-4537-b075-a1fb5c837408",
   "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>feat_1</th>\n",
       "      <th>feat_2</th>\n",
       "      <th>feat_3</th>\n",
       "      <th>feat_4</th>\n",
       "      <th>feat_5</th>\n",
       "      <th>feat_6</th>\n",
       "      <th>feat_7</th>\n",
       "      <th>feat_8</th>\n",
       "      <th>feat_9</th>\n",
       "      <th>feat_10</th>\n",
       "      <th>...</th>\n",
       "      <th>feat_84</th>\n",
       "      <th>feat_85</th>\n",
       "      <th>feat_86</th>\n",
       "      <th>feat_87</th>\n",
       "      <th>feat_88</th>\n",
       "      <th>feat_89</th>\n",
       "      <th>feat_90</th>\n",
       "      <th>feat_91</th>\n",
       "      <th>feat_92</th>\n",
       "      <th>feat_93</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61873</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61874</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61875</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61876</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61877</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>61878 rows × 93 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       feat_1  feat_2  feat_3  feat_4  feat_5  feat_6  feat_7  feat_8  feat_9  \\\n",
       "0           1       0       0       0       0       0       0       0       0   \n",
       "1           0       0       0       0       0       0       0       1       0   \n",
       "2           0       0       0       0       0       0       0       1       0   \n",
       "3           1       0       0       1       6       1       5       0       0   \n",
       "4           0       0       0       0       0       0       0       0       0   \n",
       "...       ...     ...     ...     ...     ...     ...     ...     ...     ...   \n",
       "61873       1       0       0       1       1       0       0       0       0   \n",
       "61874       4       0       0       0       0       0       0       0       0   \n",
       "61875       0       0       0       0       0       0       0       3       1   \n",
       "61876       1       0       0       0       0       0       0       0       0   \n",
       "61877       0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "       feat_10  ...  feat_84  feat_85  feat_86  feat_87  feat_88  feat_89  \\\n",
       "0            0  ...        0        1        0        0        0        0   \n",
       "1            0  ...        0        0        0        0        0        0   \n",
       "2            0  ...        0        0        0        0        0        0   \n",
       "3            1  ...       22        0        1        2        0        0   \n",
       "4            0  ...        0        1        0        0        0        0   \n",
       "...        ...  ...      ...      ...      ...      ...      ...      ...   \n",
       "61873        0  ...        0        1        0        0        0        0   \n",
       "61874        0  ...        0        0        2        0        0        2   \n",
       "61875        0  ...        0        0        3        1        0        0   \n",
       "61876        0  ...        0        0        0        0        0        1   \n",
       "61877        0  ...        0        0        0        0        0        0   \n",
       "\n",
       "       feat_90  feat_91  feat_92  feat_93  \n",
       "0            0        0        0        0  \n",
       "1            0        0        0        0  \n",
       "2            0        0        0        0  \n",
       "3            0        0        0        0  \n",
       "4            1        0        0        0  \n",
       "...        ...      ...      ...      ...  \n",
       "61873        0        0        2        0  \n",
       "61874        0        0        1        0  \n",
       "61875        0        0        0        0  \n",
       "61876        0        3       10        0  \n",
       "61877        0        0        2        0  \n",
       "\n",
       "[61878 rows x 93 columns]"
      ]
     },
     "execution_count": 224,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "id": "1c36e6cf-5206-4859-8193-7a44e034c9d5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        0\n",
       "1        0\n",
       "2        0\n",
       "3        0\n",
       "4        0\n",
       "        ..\n",
       "61873    8\n",
       "61874    8\n",
       "61875    8\n",
       "61876    8\n",
       "61877    8\n",
       "Name: target, Length: 61878, dtype: int64"
      ]
     },
     "execution_count": 225,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "id": "aac35f77-83ea-4603-a3e7-6e023b099cd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# making MLP classifier\n",
    "#x, y = make_classification(n_samples=200, n_classes=9, random_state=1, n_clusters_per_class=1, n_informative=5)\n",
    "\n",
    "# split train and test\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.5)\n",
    "\n",
    "# rescale train data to decrease sensitivity of model to the data\n",
    "sc = StandardScaler()\n",
    "scaler = sc.fit(X_train)\n",
    "trainX_scaled = scaler.transform(X_train)\n",
    "testX_scaled = scaler.transform(X_test)\n",
    "\n",
    "clf = MLPClassifier(hidden_layer_sizes=(10, 10, 10, 5), max_iter=500, activation='relu', solver='adam').fit(trainX_scaled, Y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "id": "81d0cb9d-b11f-4f7d-843b-7af9e1e7f94d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.77\n"
     ]
    }
   ],
   "source": [
    "#clf.predict_proba(X_test[:1])\n",
    "y_pred = clf.predict(testX_scaled)\n",
    "\n",
    "print('Accuracy: {:.2f}'.format(accuracy_score(Y_test, y_pred)))\n",
    "# only get 0.79 on accuracy..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "id": "b994054b-5f69-408c-944b-010dce9e6bee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.38      0.63      0.48       605\n",
      "           1       0.81      0.69      0.75      9441\n",
      "           2       0.41      0.52      0.46      3128\n",
      "           3       0.45      0.61      0.51      1022\n",
      "           4       0.94      0.96      0.95      1350\n",
      "           5       0.92      0.93      0.93      7012\n",
      "           6       0.65      0.66      0.65      1387\n",
      "           7       0.89      0.87      0.88      4318\n",
      "           8       0.86      0.81      0.84      2676\n",
      "\n",
      "    accuracy                           0.77     30939\n",
      "   macro avg       0.70      0.74      0.72     30939\n",
      "weighted avg       0.79      0.77      0.78     30939\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(classification_report(y_pred, Y_test))\n",
    "fig = plot_confusion_matrix(clf, testX_scaled, Y_test, display_labels=clf.classes_)\n",
    "fig.figure_.suptitle(\"Confusion Matrix for Otto Group Dataset\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "id": "2202d7b7-da2b-43d3-b532-0bde61517d38",
   "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": [
    "plt.plot(clf.loss_curve_)\n",
    "plt.title(\"Loss Curve\", fontsize=14)\n",
    "plt.xlabel(\"Iterations\")\n",
    "plt.ylabel(\"Cost\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b114a210-c63c-4b9d-a346-bb258ab7905a",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}