mnist-dt.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MNIST Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2020-11-21T16:20:30.785745Z",
     "iopub.status.busy": "2020-11-21T16:20:30.785040Z",
     "iopub.status.idle": "2020-11-21T16:20:32.049312Z",
     "shell.execute_reply": "2020-11-21T16:20:32.050583Z"
    },
    "papermill": {
     "duration": 1.307821,
     "end_time": "2020-11-21T16:20:32.050799",
     "exception": false,
     "start_time": "2020-11-21T16:20:30.742978",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test = pd.read_csv('test.csv')\n",
    "train = pd.read_csv('train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-21T16:20:39.364765Z",
     "iopub.status.busy": "2020-11-21T16:20:39.363384Z",
     "iopub.status.idle": "2020-11-21T16:20:39.367489Z",
     "shell.execute_reply": "2020-11-21T16:20:39.368050Z"
    },
    "papermill": {
     "duration": 0.132175,
     "end_time": "2020-11-21T16:20:39.368194",
     "exception": false,
     "start_time": "2020-11-21T16:20:39.236019",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    0\n",
       "2    1\n",
       "3    4\n",
       "4    0\n",
       "5    0\n",
       "6    7\n",
       "7    3\n",
       "8    5\n",
       "9    3\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_Train=train.drop(labels=\"label\",axis=1)\n",
    "y_Train=train[\"label\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-21T16:20:39.971144Z",
     "iopub.status.busy": "2020-11-21T16:20:39.970333Z",
     "iopub.status.idle": "2020-11-21T16:20:40.199482Z",
     "shell.execute_reply": "2020-11-21T16:20:40.198871Z"
    },
    "papermill": {
     "duration": 0.28268,
     "end_time": "2020-11-21T16:20:40.199620",
     "exception": false,
     "start_time": "2020-11-21T16:20:39.916940",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x train shape (42000, 28, 28, 1)\n",
      "x test shape (28000, 28, 28, 1)\n"
     ]
    }
   ],
   "source": [
    "x_Train=x_Train/255\n",
    "x_Test=test/255\n",
    "x_Train=x_Train.values.reshape(-1,28,28,1)\n",
    "x_Test=x_Test.values.reshape(-1,28,28,1)\n",
    "print(\"x train shape\",x_Train.shape)\n",
    "print(\"x test shape\",x_Test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-21T16:20:40.369968Z",
     "iopub.status.busy": "2020-11-21T16:20:40.368672Z",
     "iopub.status.idle": "2020-11-21T16:20:40.760553Z",
     "shell.execute_reply": "2020-11-21T16:20:40.761291Z"
    },
    "papermill": {
     "duration": 0.4404,
     "end_time": "2020-11-21T16:20:40.761468",
     "exception": false,
     "start_time": "2020-11-21T16:20:40.321068",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "X_train, X_val, ml_train, ml_val = train_test_split(x_Train, y_Train, test_size = 0.1, random_state=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-21T16:20:40.931079Z",
     "iopub.status.busy": "2020-11-21T16:20:40.930217Z",
     "iopub.status.idle": "2020-11-21T16:20:47.156713Z",
     "shell.execute_reply": "2020-11-21T16:20:47.155519Z"
    },
    "papermill": {
     "duration": 6.27151,
     "end_time": "2020-11-21T16:20:47.156852",
     "exception": false,
     "start_time": "2020-11-21T16:20:40.885342",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras.utils import to_categorical\n",
    "Y_train=to_categorical(y_Train,num_classes=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-21T16:20:47.252371Z",
     "iopub.status.busy": "2020-11-21T16:20:47.250783Z",
     "iopub.status.idle": "2020-11-21T16:20:47.642660Z",
     "shell.execute_reply": "2020-11-21T16:20:47.642039Z"
    },
    "papermill": {
     "duration": 0.443467,
     "end_time": "2020-11-21T16:20:47.642788",
     "exception": false,
     "start_time": "2020-11-21T16:20:47.199321",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape (37800, 28, 28, 1)\n",
      "x_val shape (4200, 28, 28, 1)\n",
      "y_train shape (37800, 10)\n",
      "y_val shape (4200, 10)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_val, Y_train, Y_val = train_test_split(x_Train, Y_train, test_size = 0.1, random_state=2)\n",
    "print(\"x_train shape\",X_train.shape)\n",
    "print(\"x_val shape\",X_val.shape)\n",
    "print(\"y_train shape\",Y_train.shape)\n",
    "print(\"y_val shape\",Y_val.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-21T16:42:37.340018Z",
     "iopub.status.busy": "2020-11-21T16:42:37.339164Z",
     "iopub.status.idle": "2020-11-21T16:42:50.824506Z",
     "shell.execute_reply": "2020-11-21T16:42:50.823889Z"
    },
    "papermill": {
     "duration": 16.313002,
     "end_time": "2020-11-21T16:42:50.824623",
     "exception": false,
     "start_time": "2020-11-21T16:42:34.511621",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9302380952380952\n"
     ]
    }
   ],
   "source": [
    "from sklearn import tree\n",
    "clf = tree.DecisionTreeClassifier()\n",
    "clf = clf.fit(train_x, ml_train)\n",
    "y_pred = clf.predict(test_x)\n",
    "\n",
    "from sklearn.metrics import confusion_matrix\n",
    "mat = confusion_matrix(ml_val,y_pred)\n",
    "fig = plt.figure(figsize=(8,8))\n",
    "\n",
    "sns.set(font_scale=2)\n",
    "sns.heatmap(mat.T, annot=True, cbar = False, annot_kws={'size': 16}, fmt='.0f')\n",
    "    \n",
    "plt.title(\"Confusion Matrix\")\n",
    "plt.xlabel('Predicted Values')\n",
    "plt.ylabel('True Values')\n",
    "plt.show()\n",
    "\n",
    "from sklearn.metrics import accuracy_score\n",
    "print(\"Accuracy: \" + accuracy_score(ml_val,y_pred).__str__())"
   ]
  }
 ],
 "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": "3.8.12"
  },
  "papermill": {
   "duration": 1390.189742,
   "end_time": "2020-11-21T16:43:35.090781",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2020-11-21T16:20:24.901039",
   "version": "2.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}