section2_Cuisine_prediction.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "polyphonic-premises",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import sklearn\n",
    "import matplotlib\n",
    "import xgboost\n",
    "#import graphviz\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.tree import DecisionTreeClassifier, export_graphviz\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, r2_score\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.decomposition import KernelPCA\n",
    "\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import LinearSVC, SVC\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import confusion_matrix, log_loss\n",
    "from pandas.api.types import CategoricalDtype\n",
    "\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "from xgboost.sklearn import XGBClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.dummy import DummyClassifier\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "alternative-walter",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "* Pandas 0.25.3\n",
      "* sklearn 0.23.2\n",
      "* seaborn 0.10.0\n",
      "* matplotlib 3.3.2\n",
      "* numpy 1.19.2\n",
      "* xgboost 1.19.2\n"
     ]
    }
   ],
   "source": [
    "print('* Pandas', pd.__version__)\n",
    "print('* sklearn', sklearn.__version__)\n",
    "print('* seaborn', sns.__version__)\n",
    "print('* matplotlib', matplotlib.__version__)\n",
    "print('* numpy', np.__version__)\n",
    "print('* xgboost', np.__version__)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "rising-amazon",
   "metadata": {
    "scrolled": true
   },
   "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>'acorn squash'</th>\n",
       "      <th>adobo</th>\n",
       "      <th>'african birdseye chile pepper'</th>\n",
       "      <th>ale</th>\n",
       "      <th>'aleppo pepper'</th>\n",
       "      <th>'alfalfa sprouts'</th>\n",
       "      <th>'alfredo sauce'</th>\n",
       "      <th>allspice</th>\n",
       "      <th>almond</th>\n",
       "      <th>'almond butter'</th>\n",
       "      <th>...</th>\n",
       "      <th>yeast</th>\n",
       "      <th>'yellow curry paste'</th>\n",
       "      <th>'yellow food coloring'</th>\n",
       "      <th>'yellow split pea'</th>\n",
       "      <th>'yellow squash'</th>\n",
       "      <th>yogurt</th>\n",
       "      <th>zaatar</th>\n",
       "      <th>zest</th>\n",
       "      <th>zucchini</th>\n",
       "      <th>cuisine</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</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>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>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>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>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>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</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>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>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>...</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>4231</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>0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4232</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>0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4233</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>0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4234</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>0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4235</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>0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4236 rows × 710 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      'acorn squash'  adobo  'african birdseye chile pepper'  ale  \\\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                  0      0                                0    0   \n",
       "...              ...    ...                              ...  ...   \n",
       "4231               0      0                                0    0   \n",
       "4232               0      0                                0    0   \n",
       "4233               0      0                                0    0   \n",
       "4234               0      0                                0    0   \n",
       "4235               0      0                                0    0   \n",
       "\n",
       "      'aleppo pepper'  'alfalfa sprouts'  'alfredo sauce'  allspice  almond  \\\n",
       "0                   0                  0                0         0       0   \n",
       "1                   0                  0                0         0       0   \n",
       "2                   0                  0                0         0       0   \n",
       "3                   0                  0                0         0       0   \n",
       "4                   0                  0                0         0       0   \n",
       "...               ...                ...              ...       ...     ...   \n",
       "4231                0                  0                0         0       0   \n",
       "4232                0                  0                0         0       0   \n",
       "4233                0                  0                0         0       0   \n",
       "4234                0                  0                0         0       0   \n",
       "4235                0                  0                0         0       0   \n",
       "\n",
       "      'almond butter'  ...  yeast  'yellow curry paste'  \\\n",
       "0                   0  ...      0                     0   \n",
       "1                   0  ...      0                     0   \n",
       "2                   0  ...      0                     0   \n",
       "3                   0  ...      0                     0   \n",
       "4                   0  ...      0                     0   \n",
       "...               ...  ...    ...                   ...   \n",
       "4231                0  ...      0                     0   \n",
       "4232                0  ...      0                     0   \n",
       "4233                0  ...      0                     0   \n",
       "4234                0  ...      0                     0   \n",
       "4235                0  ...      0                     0   \n",
       "\n",
       "      'yellow food coloring'  'yellow split pea'  'yellow squash'  yogurt  \\\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                          0                   0                0       0   \n",
       "...                      ...                 ...              ...     ...   \n",
       "4231                       0                   0                0       0   \n",
       "4232                       0                   0                0       0   \n",
       "4233                       0                   0                0       0   \n",
       "4234                       0                   0                0       0   \n",
       "4235                       0                   0                0       0   \n",
       "\n",
       "      zaatar  zest  zucchini  cuisine  \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          0     0         0        0  \n",
       "...      ...   ...       ...      ...  \n",
       "4231       0     0         0       11  \n",
       "4232       0     0         0       11  \n",
       "4233       0     0         0       11  \n",
       "4234       0     0         0       11  \n",
       "4235       0     0         0       11  \n",
       "\n",
       "[4236 rows x 710 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the dataset\n",
    "data_path = os.path.join(os.getcwd(),  'recipes.csv')\n",
    "data = pd.read_csv(data_path)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "amino-transcription",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11    353\n",
       "7     353\n",
       "3     353\n",
       "10    353\n",
       "6     353\n",
       "2     353\n",
       "9     353\n",
       "5     353\n",
       "1     353\n",
       "8     353\n",
       "4     353\n",
       "0     353\n",
       "Name: cuisine, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['cuisine'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "little-acting",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X shape: (4236, 709)\n",
      "y shape: (4236,)\n"
     ]
    }
   ],
   "source": [
    "data_clean = data.copy(deep=True) # Make a copy of the DataFrame\n",
    "\n",
    "# Extract (and print the shape of) X\n",
    "X = data_clean.drop('cuisine', axis=1)\n",
    "column_names = X.columns\n",
    "# Convert to Numpy Array\n",
    "X = X.values\n",
    "print('X shape: {}'.format(X.shape))\n",
    "\n",
    "# Extract the Target y (and print its shape)\n",
    "data_clean['cuisine'] = data_clean['cuisine'].astype('category') # Convert using astype(...) method\n",
    "#convert the categorical values into a one-hot-encoding\n",
    "#y = pd.get_dummies(data_clean['cuisine']) # Target vector\n",
    "# Convert to Numpy Array\n",
    "y = data_clean['cuisine']\n",
    "#y = y.values\n",
    "print('y shape: {}'.format(np.shape(y)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "small-sheet",
   "metadata": {},
   "source": [
    "### ========== simple PCA =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "clean-galaxy",
   "metadata": {},
   "outputs": [],
   "source": [
    "pca = PCA(n_components=2, random_state=0)\n",
    "X_2d = pca.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "faced-notice",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "class_categories = ['Chinese', 'English', 'French', 'German', 'Greek', 'Indian', 'Italian', 'Japanese', 'Mexican', 'Moroccan', 'Spanish', 'Thai']\n",
    "class_labels = []\n",
    "for i in range(0,12):\n",
    "    class_labels.append(i)\n",
    "\n",
    "# PCA\n",
    "pca = PCA(n_components=2)\n",
    "X_2d = pca.fit_transform(X)\n",
    "\n",
    "y_ = data['cuisine'].values\n",
    "\n",
    "#X_2d_plot = np.concatenate((X_2d,y_),axis=1)\n",
    "\n",
    "plt.figure(figsize=(12,8))\n",
    "for label, cat in zip(class_labels, class_categories):\n",
    "    plt.scatter(X_2d[y_ == label, 0], X_2d[y_ == label, 1], alpha=.5, lw=2, label=cat)\n",
    "plt.axis('equal')\n",
    "plt.legend(loc='center left', scatterpoints=3, bbox_to_anchor=[1.01, 0.5])\n",
    "plt.title('Labelled data in PCA space')\n",
    "plt.xlabel('PC1')\n",
    "plt.ylabel('PC2')\n",
    "top_plot = plt.gca()\n",
    "\n",
    "plt.savefig('pca_no_kernel.pdf', bbox_inches = 'tight')\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "automotive-female",
   "metadata": {},
   "source": [
    "## ========== 5-fold Cross Validation ==========\n",
    "See above, the train acc is much higher than test acc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "greenhouse-referral",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "StratifiedKFold(n_splits=5, random_state=0, shuffle=True) 5\n"
     ]
    }
   ],
   "source": [
    "#  split off stratified test data (20%) by preserving the percentage of samples for each class.\n",
    "skf = StratifiedKFold(n_splits=5, random_state=0, shuffle=True)\n",
    "print(skf, skf.get_n_splits(X, y))\n",
    "for train_index, test_index in skf.split(X, y):\n",
    "    X_fold, X_cvtest = X[train_index], X[test_index]\n",
    "    y_fold, y_cvtest = y[train_index], y[test_index]\n",
    "    \n",
    "# 20% for test (X_cvtest, y_cvtest) , 80% for train and validate (X_fold, y_fold)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "associate-radio",
   "metadata": {},
   "outputs": [],
   "source": [
    "#print(\"TRAIN:\", train_index, \"TEST:\", test_index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "written-gibraltar",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3389, 709)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_fold.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "falling-stylus",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(847, 709)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_cvtest.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "realistic-genome",
   "metadata": {},
   "source": [
    "## ========== Baseline =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "quantitative-hometown",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification accuracy on the validation set:\n",
      "'Most-frequent' dummy classifier: 0.0826\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0.08259587, 0.08259587, 0.08259587, 0.08259587, 0.08271787])"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dclf = DummyClassifier(strategy='most_frequent')\n",
    "#pipe = Pipeline([('scaler', StandardScaler()), ('dclf', dclf)])\n",
    "\n",
    "score_dclf = cross_val_score(dclf, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "\n",
    "\n",
    "\n",
    "print(\"Classification accuracy on the validation set:\")\n",
    "print(\"'Most-frequent' dummy classifier: {0:.4f}\".format(score_dclf.mean()))\n",
    "      \n",
    "\n",
    "score_dclf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "approximate-specification",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification accuracy on the validation set:\n",
      "'uniform' dummy classifier: 0.0841\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0.06342183, 0.06932153, 0.08259587, 0.10324484, 0.10192024])"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dclf = DummyClassifier(strategy='uniform')\n",
    "#pipe = Pipeline([('scaler', StandardScaler()), ('dclf', dclf)])\n",
    "\n",
    "score_dclf = cross_val_score(dclf, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "\n",
    "\n",
    "\n",
    "print(\"Classification accuracy on the validation set:\")\n",
    "print(\"'uniform' dummy classifier: {0:.4f}\".format(score_dclf.mean()))\n",
    "score_dclf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "equivalent-blues",
   "metadata": {},
   "source": [
    "### ========== 1. Linear SVM (5-fold CV) =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "affiliated-small",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C = 0.01, Linear SVC classification accuracy on validation set: 0.753\n",
      "C = 0.02, Linear SVC classification accuracy on validation set: 0.764\n",
      "C = 0.05, Linear SVC classification accuracy on validation set: 0.770\n",
      "C = 0.08, Linear SVC classification accuracy on validation set: 0.771\n",
      "C = 0.1, Linear SVC classification accuracy on validation set: 0.770\n",
      "C = 0.5, Linear SVC classification accuracy on validation set: 0.750\n",
      "C = 1.0, Linear SVC classification accuracy on validation set: 0.739\n",
      "C = 1.5, Linear SVC classification accuracy on validation set: 0.731\n"
     ]
    }
   ],
   "source": [
    "# classification accuracy with different C (Penalty parameter C of the error term)\n",
    "C1 = [0.01, 0.02, 0.05, 0.08, 0.1, 0.5, 1.0, 1.5]\n",
    "scores_svc_linear = []\n",
    "for c in C1:\n",
    "    cv_svc_linear = LinearSVC(C=c, random_state = 0)\n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score_svc_linear = cross_val_score(cv_svc_linear, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores_svc_linear.append(score_svc_linear)\n",
    "    print('C = {}, Linear SVC classification accuracy on validation set: {:.3f}'.format(c, score_svc_linear.mean()))\n",
    "    \n",
    "scores_svc_linear = np.array(scores_svc_linear)\n",
    "mean_acc_lsvc = scores_svc_linear.mean(axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "innovative-tuesday",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "#plt.xscale('Penalty parameter C of the error term')\n",
    "plt.plot(C1, mean_acc_lsvc, marker='*', color = 'magenta',alpha=.5, ms=8)\n",
    "plt.xlabel(\"Regularisation Parameter (C)\")\n",
    "plt.ylabel(\"Mean Cross-validated Classification Accuracy\")\n",
    "plt.style.use('ggplot')\n",
    "#plt.savefig('2.5.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "accepting-field",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final Linear SVC classification accuracy on training set: 0.883\n",
      "Final Linear SVC classification accuracy on test set: 0.796\n"
     ]
    }
   ],
   "source": [
    "# final model using proper parameters:\n",
    "# c = 0.08\n",
    "cv_svc_linear = LinearSVC(C=0.08, random_state = 0)\n",
    "cv_svc_linear.fit(X_fold, y_fold)\n",
    "print('Final Linear SVC classification accuracy on training set: {:.3f}'.format(cv_svc_linear.score(X_fold, y_fold)))\n",
    "print('Final Linear SVC classification accuracy on test set: {:.3f}'.format(cv_svc_linear.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "level-supplement",
   "metadata": {},
   "source": [
    "### ==========2. Kernel SVM (5-fold CV) =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "later-blogger",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C = 0.5, rbf SVC classification accuracy on validation set: 0.738\n",
      "C = 1.0, rbf SVC classification accuracy on validation set: 0.752\n",
      "C = 1.5, rbf SVC classification accuracy on validation set: 0.758\n",
      "C = 2.0, rbf SVC classification accuracy on validation set: 0.764\n",
      "C = 3.0, rbf SVC classification accuracy on validation set: 0.760\n",
      "C = 3.5, rbf SVC classification accuracy on validation set: 0.758\n",
      "C = 4.0, rbf SVC classification accuracy on validation set: 0.757\n",
      "C = 4.5, rbf SVC classification accuracy on validation set: 0.759\n",
      "C = 5.0, rbf SVC classification accuracy on validation set: 0.758\n"
     ]
    }
   ],
   "source": [
    "# classification accuracy with different C (Penalty parameter C of the error term)\n",
    "C2 = [0.5, 1.0, 1.5, 2.0, 3.0, 3.5, 4.0, 4.5, 5.0]\n",
    "scores = []\n",
    "for c in C2:\n",
    "    classifier = SVC(C=c, kernel='rbf', random_state = 0)\n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('C = {}, rbf SVC classification accuracy on validation set: {:.3f}'.format(c, score.mean()))\n",
    "    \n",
    "scores = np.array(scores)\n",
    "mean_acc = scores.mean(axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "massive-bruce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "#plt.xscale('log')\n",
    "plt.plot(C2, mean_acc, marker='*', color = 'magenta',alpha=.5, ms=8)\n",
    "plt.xlabel(\"Regularisation Parameter (C)\")\n",
    "plt.ylabel(\"Mean Cross-validated Classification Accuracy\")\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "serious-married",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final rbf SVC classification accuracy on training set: 0.976\n",
      "Final rbf SVC classification accuracy on test set: 0.787\n"
     ]
    }
   ],
   "source": [
    "# final model using proper parameters:\n",
    "# c = 2.0\n",
    "cv_svc_rbf =SVC(C=2.0, kernel='rbf', random_state = 0)\n",
    "cv_svc_rbf.fit(X_fold, y_fold)\n",
    "print('Final rbf SVC classification accuracy on training set: {:.3f}'.format(cv_svc_rbf.score(X_fold, y_fold)))\n",
    "print('Final rbf SVC classification accuracy on test set: {:.3f}'.format(cv_svc_rbf.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "historic-luxembourg",
   "metadata": {},
   "source": [
    "### ==========3. Logistic Regression  (5-fold CV)  =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "unnecessary-unknown",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C = 0.01, rbf SVC classification accuracy on validation set: 0.708\n",
      "C = 0.02, rbf SVC classification accuracy on validation set: 0.718\n",
      "C = 0.05, rbf SVC classification accuracy on validation set: 0.734\n",
      "C = 0.08, rbf SVC classification accuracy on validation set: 0.742\n",
      "C = 0.1, rbf SVC classification accuracy on validation set: 0.742\n",
      "C = 0.5, rbf SVC classification accuracy on validation set: 0.763\n",
      "C = 1.0, rbf SVC classification accuracy on validation set: 0.760\n",
      "C = 1.5, rbf SVC classification accuracy on validation set: 0.761\n"
     ]
    }
   ],
   "source": [
    "# classification accuracy with different C (Penalty parameter C of the error term)\n",
    "C1 = [0.01, 0.02, 0.05, 0.08, 0.1, 0.5, 1.0, 1.5]\n",
    "scores = []\n",
    "for c in C1:\n",
    "    classifier =  LogisticRegression(C=c, solver='sag', random_state = 0) \n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('C = {}, rbf SVC classification accuracy on validation set: {:.3f}'.format(c, score.mean()))\n",
    "    \n",
    "scores = np.array(scores)\n",
    "mean_acc = scores.mean(axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "elegant-chemical",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.884\n",
      "Final lr classification accuracy on test set: 0.800\n"
     ]
    }
   ],
   "source": [
    "# final model using proper parameters:\n",
    "# c = 0.5\n",
    "cv_lr =LogisticRegression(C=0.5, solver='sag', random_state = 0) \n",
    "cv_lr.fit(X_fold, y_fold)\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(cv_lr.score(X_fold, y_fold)))\n",
    "print('Final lr classification accuracy on test set: {:.3f}'.format(cv_lr.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "foster-cutting",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " Confusion matrix \n",
      "[[51  0  0  0  0  0  0 13  1  0  0  5]\n",
      " [ 0 49  7  4  2  2  2  1  1  0  3  0]\n",
      " [ 0  3 51  9  0  1  3  0  0  0  3  0]\n",
      " [ 0  5  4 58  1  0  1  0  0  1  1  0]\n",
      " [ 0  4  3  1 53  1  3  0  0  3  3  0]\n",
      " [ 1  2  0  0  0 63  0  1  0  1  0  2]\n",
      " [ 0  1  4  1  5  0 55  0  0  0  5  0]\n",
      " [13  0  0  1  0  1  0 52  1  0  0  2]\n",
      " [ 0  1  0  2  2  1  0  0 65  0  0  0]\n",
      " [ 0  0  1  0  0  4  0  0  1 62  3  0]\n",
      " [ 0  4  5  0  1  0  1  0  0  1 58  0]\n",
      " [ 2  0  0  0  0  4  0  3  1  0  0 61]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "import seaborn as sns\n",
    "\n",
    "y_pred = cv_lr.predict(X_cvtest)\n",
    "cm_test = confusion_matrix(y_cvtest, y_pred)\n",
    "np.set_printoptions(precision=2)  # only print two significant digits\n",
    "print('\\n Confusion matrix \\n{}'.format(cm_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "average-yahoo",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(cm, class_labels=None):\n",
    "    \"\"\"Plots a confusion matrix using seaborn's heatmap function\n",
    "    \n",
    "    Columns and rows are labelled with the strings provided in class_labels.\n",
    "    \n",
    "    Parameters\n",
    "    ----------\n",
    "    cm: array-like\n",
    "        contains the confusion matrix\n",
    "        \n",
    "    class_labels: array-like, optional\n",
    "        contains the string labels\n",
    "            \n",
    "    \"\"\"\n",
    "    \n",
    "    # check whether we have count data or not\n",
    "    if issubclass(cm.dtype.type, np.integer):\n",
    "        fmt = 'd'\n",
    "    else:\n",
    "        fmt = '.2f'\n",
    "    \n",
    "    # Your code goes here\n",
    "    \n",
    "    if class_labels is not None:\n",
    "        ax=sns.heatmap(cm, cmap='viridis',xticklabels=class_labels, yticklabels=class_labels,\\\n",
    "                    annot=True, annot_kws={\"fontsize\":9},  fmt=fmt)  # controls the display of the numbers\n",
    "    else:\n",
    "        sns.heatmap(cm, annot=True, annot_kws={\"fontsize\":9},  fmt=fmt)\n",
    "    \n",
    "    #plt.rcParams['figure.figsize']=9,9\n",
    "    ax.figure.set_size_inches(10,6)\n",
    "    plt.ylabel('True label', fontweight='bold')\n",
    "    plt.xlabel('Predicted label', fontweight='bold')\n",
    "    \n",
    "    # you can change the appearance of the figure with lower-level matplotlib commands\n",
    "    # here we rotate the labels on the x-axis\n",
    "    plt.setp(plt.gca().get_xticklabels(), rotation=45, ha=\"right\",\n",
    "         rotation_mode=\"anchor\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "regulation-phoenix",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "class_categories = ['Chinese', 'English', 'French', 'German', 'Greek', 'Indian', 'Italian', 'Japanese', 'Mexican', 'Moroccan', 'Spanish', 'Thai']\n",
    "plot_confusion_matrix(cm_test, class_labels=class_categories)\n",
    "plt.savefig('confu_ma.pdf', bbox_inches = 'tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abandoned-parallel",
   "metadata": {},
   "source": [
    "### ==========4. Decision Tree  (5-fold CV)  =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "august-crime",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C = 10, decision tree classification accuracy on validation set: 0.511\n",
      "C = 15, decision tree classification accuracy on validation set: 0.560\n",
      "C = 20, decision tree classification accuracy on validation set: 0.570\n",
      "C = 25, decision tree classification accuracy on validation set: 0.580\n",
      "C = 30, decision tree classification accuracy on validation set: 0.591\n",
      "C = 40, decision tree classification accuracy on validation set: 0.587\n",
      "C = 45, decision tree classification accuracy on validation set: 0.593\n",
      "C = 50, decision tree classification accuracy on validation set: 0.595\n",
      "C = 55, decision tree classification accuracy on validation set: 0.595\n",
      "C = 60, decision tree classification accuracy on validation set: 0.595\n"
     ]
    }
   ],
   "source": [
    "# try different max depth\n",
    "C3 = [10, 15, 20, 25, 30, 40, 45, 50, 55, 60]\n",
    "scores = []\n",
    "for c in C3:\n",
    "    classifier =  DecisionTreeClassifier(max_depth = c, random_state = 0) \n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('C = {}, decision tree classification accuracy on validation set: {:.3f}'.format(c, score.mean()))\n",
    "    \n",
    "scores = np.array(scores)\n",
    "mean_acc = scores.mean(axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "whole-football",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.995\n",
      "Final lr classification accuracy on test set: 0.619\n"
     ]
    }
   ],
   "source": [
    "# final model using proper parameters:\n",
    "# c = 40\n",
    "dt = DecisionTreeClassifier(max_depth=45, random_state=0)\n",
    "dt.fit(X_fold, y_fold)\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(dt.score(X_fold, y_fold)))\n",
    "print('Final lr classification accuracy on test set: {:.3f}'.format(dt.score(X_cvtest, y_cvtest)))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "therapeutic-shadow",
   "metadata": {},
   "source": [
    "### ==========5. KNN  (5-fold CV)  =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "adequate-bottom",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 5, rbf SVC classification accuracy on validation set: 0.620\n",
      "k = 8, rbf SVC classification accuracy on validation set: 0.629\n",
      "k = 11, rbf SVC classification accuracy on validation set: 0.629\n",
      "k = 14, rbf SVC classification accuracy on validation set: 0.638\n",
      "k = 17, rbf SVC classification accuracy on validation set: 0.636\n",
      "k = 20, rbf SVC classification accuracy on validation set: 0.644\n",
      "k = 23, rbf SVC classification accuracy on validation set: 0.640\n",
      "k = 26, rbf SVC classification accuracy on validation set: 0.643\n",
      "k = 29, rbf SVC classification accuracy on validation set: 0.640\n"
     ]
    }
   ],
   "source": [
    "# try different number of neighbors k\n",
    "scores = []\n",
    "for k in np.arange(5, 30, 3) :\n",
    "    classifier =  KNeighborsClassifier(n_neighbors= k )\n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('k = {}, rbf SVC classification accuracy on validation set: {:.3f}'.format(k, score.mean()))\n",
    "    \n",
    "scores = np.array(scores)\n",
    "mean_acc = scores.mean(axis = 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "appointed-bryan",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN classification (k=20) accuracy on test set: 0.665\n"
     ]
    }
   ],
   "source": [
    "# final model using proper parameters:\n",
    "# k = 20\n",
    "knn = KNeighborsClassifier(n_neighbors=20)\n",
    "knn.fit(X_fold, y_fold)\n",
    "print('KNN classification (k=20) accuracy on test set: {:.3f}'.format(knn.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "hazardous-tension",
   "metadata": {},
   "source": [
    "### ==========6. XGBoost  (5-fold CV)  =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "scenic-boundary",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[23:22:32] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:22:52] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:23:13] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:23:32] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:23:53] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "k = 29, rbf SVC classification accuracy on validation set: 0.736\n",
      "[23:24:13] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:24:43] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:25:15] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:25:43] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:26:13] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "k = 29, rbf SVC classification accuracy on validation set: 0.734\n",
      "[23:26:40] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:27:19] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:27:58] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:28:53] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:29:43] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "k = 29, rbf SVC classification accuracy on validation set: 0.740\n",
      "[23:30:31] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:31:25] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:32:19] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:33:12] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:34:08] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "k = 29, rbf SVC classification accuracy on validation set: 0.739\n",
      "[23:35:01] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:36:03] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:36:56] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:37:54] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "[23:38:51] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "k = 29, rbf SVC classification accuracy on validation set: 0.736\n"
     ]
    }
   ],
   "source": [
    "scores = []\n",
    "for c in [20, 30, 40, 50, 60] :\n",
    "    classifier =  XGBClassifier(\n",
    "                        n_estimators=c,#三十棵树\n",
    "                        learning_rate =0.3,\n",
    "                        max_depth=10,\n",
    "                        min_child_weight=1,\n",
    "                        #gamma=0.3,\n",
    "                        subsample=0.8,\n",
    "                        colsample_bytree=0.8,\n",
    "                        #objective= 'multi:softmax',\n",
    "                        num_class=12,\n",
    "                        #nthread=12,\n",
    "                        #scale_pos_weight=1,\n",
    "                        reg_lambda=1,\n",
    "                        alpha=0.3,\n",
    "                        seed=0,\n",
    "                        use_label_encoder=False\n",
    "                        )\n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('k = {}, rbf SVC classification accuracy on validation set: {:.3f}'.format(k, score.mean()))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "eight-plasma",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[23:44:33] WARNING: /Users/travis/build/dmlc/xgboost/src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'multi:softprob' was changed from 'merror' to 'mlogloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "Final lr classification accuracy on training set: 0.981\n",
      "KNN classification (k=20) accuracy on test set: 0.790\n"
     ]
    }
   ],
   "source": [
    "xgb = XGBClassifier(\n",
    "    n_estimators=40,#三十棵树\n",
    "    learning_rate =0.3,\n",
    "    max_depth=10,\n",
    "    min_child_weight=1,\n",
    "    #gamma=0.3,\n",
    "    subsample=0.8,\n",
    "    colsample_bytree=0.8,\n",
    "    #objective= 'multi:softmax',\n",
    "    num_class=12,\n",
    "    #nthread=12,\n",
    "    #scale_pos_weight=1,\n",
    "    reg_lambda=1,\n",
    "    alpha=0.3,\n",
    "    seed=0,\n",
    "    use_label_encoder=False\n",
    "    )\n",
    "\n",
    "xgb.fit(X_fold, y_fold)\n",
    "\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(xgb.score(X_fold, y_fold)))\n",
    "print('KNN classification (k=20) accuracy on test set: {:.3f}'.format(xgb.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "standing-commons",
   "metadata": {},
   "source": [
    "### ==========7. MLP  (5-fold CV)  =========="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "extra-proposal",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.970\n",
      "Final lr classification accuracy on test set: 0.754\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (300) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "scaler = StandardScaler()\n",
    "scaler.fit(X_fold)\n",
    "X_fold_sca = scaler.transform(X_fold)\n",
    "X_cvtest_sca = scaler.transform(X_cvtest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "id": "descending-camcorder",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.979\n",
      "Final lr classification accuracy on test set: 0.750\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (20) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "nn = MLPClassifier(hidden_layer_sizes=(128, 24), activation='relu', solver='adam', alpha=0.001, batch_size=300, random_state=0, max_iter=20)\n",
    "nn.fit(X_fold_sca, y_fold)\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(nn.score(X_fold_sca, y_fold)))\n",
    "print('Final lr classification accuracy on test set: {:.3f}'.format(nn.score(X_cvtest_sca, y_cvtest)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "id": "novel-shield",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.942\n",
      "Final lr classification accuracy on test set: 0.791\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (10) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "nn = MLPClassifier(hidden_layer_sizes=(512, 128, 24), activation='relu', solver='adam', alpha=0.0001, batch_size=300, random_state=0, max_iter=10)\n",
    "nn.fit(X_fold, y_fold)\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(nn.score(X_fold, y_fold)))\n",
    "print('Final lr classification accuracy on test set: {:.3f}'.format(nn.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "id": "forty-culture",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.982\n",
      "Final lr classification accuracy on test set: 0.796\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "nn = MLPClassifier(hidden_layer_sizes=(512, 128, 24), activation='relu', solver='sgd', \n",
    "                   alpha=0.0001, batch_size=200, random_state=0, max_iter=110,\n",
    "                  learning_rate_init=0.005)\n",
    "nn.fit(X_fold, y_fold)\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(nn.score(X_fold, y_fold)))\n",
    "print('Final lr classification accuracy on test set: {:.3f}'.format(nn.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "id": "cleared-inspection",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.970\n",
      "Final lr classification accuracy on test set: 0.737\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "nn = MLPClassifier(hidden_layer_sizes=(512, 128, 24), activation='relu', solver='sgd', \n",
    "                   alpha=0.0001, batch_size=200, random_state=0, max_iter=110,\n",
    "                  learning_rate_init=0.001)\n",
    "nn.fit(X_fold_sca, y_fold)\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(nn.score(X_fold_sca, y_fold)))\n",
    "print('Final lr classification accuracy on test set: {:.3f}'.format(nn.score(X_cvtest_sca, y_cvtest)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "id": "dental-intermediate",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (90) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (90) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (90) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (90) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (90) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 90,mlp classification accuracy on validation set: 0.750\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 110,mlp classification accuracy on validation set: 0.749\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (130) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (130) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (130) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (130) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (130) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 130,mlp classification accuracy on validation set: 0.746\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (150) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (150) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (150) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (150) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 150,mlp classification accuracy on validation set: 0.747\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (150) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "scores = []\n",
    "for a in [90, 110, 130, 150] :\n",
    "    classifier = MLPClassifier(hidden_layer_sizes=(512, 128, 24), activation='relu', solver='sgd', \n",
    "                   alpha=0.0001, batch_size=200, random_state=0, max_iter=a,\n",
    "                  learning_rate_init=0.005)\n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('k = {},mlp classification accuracy on validation set: {:.3f}'.format(a, score.mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "id": "ethical-adams",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (50) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (50) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (50) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (50) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (50) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 50,mlp classification accuracy on validation set: 0.737\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (60) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (60) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (60) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (60) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (60) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 60,mlp classification accuracy on validation set: 0.746\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (70) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (70) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (70) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (70) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (70) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 70,mlp classification accuracy on validation set: 0.750\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (80) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (80) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (80) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (80) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 80,mlp classification accuracy on validation set: 0.749\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (80) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "scores = []\n",
    "for a in [50, 60, 70, 80] :\n",
    "    classifier = MLPClassifier(hidden_layer_sizes=(512, 128, 24), activation='relu', solver='sgd', \n",
    "                   alpha=0.0001, batch_size=200, random_state=0, max_iter=a,\n",
    "                  learning_rate_init=0.005)\n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('k = {},mlp classification accuracy on validation set: {:.3f}'.format(a, score.mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "id": "traditional-segment",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (160) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (160) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (160) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (160) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (160) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 160,mlp classification accuracy on validation set: 0.745\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (180) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (180) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (180) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (180) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (180) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 180,mlp classification accuracy on validation set: 0.744\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 200,mlp classification accuracy on validation set: 0.744\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (250) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (250) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (250) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n",
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (250) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k = 250,mlp classification accuracy on validation set: 0.742\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (250) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "scores = []\n",
    "for a in [160, 180, 200, 250] :\n",
    "    classifier = MLPClassifier(hidden_layer_sizes=(512, 128, 24), activation='relu', solver='sgd', \n",
    "                   alpha=0.0001, batch_size=200, random_state=0, max_iter=a,\n",
    "                  learning_rate_init=0.005)\n",
    "    # Returns the mean accuracy on the given test data and labels.\n",
    "    score = cross_val_score(classifier, X_fold, y_fold, cv=StratifiedKFold(n_splits=5, shuffle = True, random_state=0))\n",
    "    scores.append(score)\n",
    "    print('k = {},mlp classification accuracy on validation set: {:.3f}'.format(a, score.mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "id": "stuck-membership",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final lr classification accuracy on training set: 0.982\n",
      "Final lr classification accuracy on test set: 0.796\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/zhangyijin/anaconda3/envs/DME/lib/python3.7/site-packages/sklearn/neural_network/_multilayer_perceptron.py:585: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (110) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    }
   ],
   "source": [
    "# final\n",
    "nn = MLPClassifier(hidden_layer_sizes=(512, 128, 24), activation='relu', solver='sgd', \n",
    "                   alpha=0.0001, batch_size=200, random_state=0, max_iter=110,\n",
    "                  learning_rate_init=0.005)\n",
    "nn.fit(X_fold, y_fold)\n",
    "print('Final lr classification accuracy on training set: {:.3f}'.format(nn.score(X_fold, y_fold)))\n",
    "print('Final lr classification accuracy on test set: {:.3f}'.format(nn.score(X_cvtest, y_cvtest)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "mental-emphasis",
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}