2layerdigitrecognizer.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2192ff27",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:15.576559Z",
     "iopub.status.busy": "2023-06-28T14:51:15.576204Z",
     "iopub.status.idle": "2023-06-28T14:51:17.965899Z",
     "shell.execute_reply": "2023-06-28T14:51:17.964547Z"
    },
    "papermill": {
     "duration": 2.397523,
     "end_time": "2023-06-28T14:51:17.968378",
     "exception": false,
     "start_time": "2023-06-28T14:51:15.570855",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "from matplotlib import pyplot as plt # visualization\n",
    "\n",
    "data = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "46c162f5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:17.976935Z",
     "iopub.status.busy": "2023-06-28T14:51:17.976576Z",
     "iopub.status.idle": "2023-06-28T14:51:18.010338Z",
     "shell.execute_reply": "2023-06-28T14:51:18.009301Z"
    },
    "papermill": {
     "duration": 0.040498,
     "end_time": "2023-06-28T14:51:18.012538",
     "exception": false,
     "start_time": "2023-06-28T14:51:17.972040",
     "status": "completed"
    },
    "tags": []
   },
   "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>label</th>\n",
       "      <th>pixel0</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel774</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>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>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>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",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 785 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   label  pixel0  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  \\\n",
       "0      1       0       0       0       0       0       0       0       0   \n",
       "1      0       0       0       0       0       0       0       0       0   \n",
       "2      1       0       0       0       0       0       0       0       0   \n",
       "3      4       0       0       0       0       0       0       0       0   \n",
       "4      0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   pixel8  ...  pixel774  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "0       0  ...         0         0         0         0         0         0   \n",
       "1       0  ...         0         0         0         0         0         0   \n",
       "2       0  ...         0         0         0         0         0         0   \n",
       "3       0  ...         0         0         0         0         0         0   \n",
       "4       0  ...         0         0         0         0         0         0   \n",
       "\n",
       "   pixel780  pixel781  pixel782  pixel783  \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",
       "[5 rows x 785 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head() #view the first five rows of the pandas dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5625635c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:18.021583Z",
     "iopub.status.busy": "2023-06-28T14:51:18.021218Z",
     "iopub.status.idle": "2023-06-28T14:51:18.431611Z",
     "shell.execute_reply": "2023-06-28T14:51:18.430421Z"
    },
    "papermill": {
     "duration": 0.417663,
     "end_time": "2023-06-28T14:51:18.434082",
     "exception": false,
     "start_time": "2023-06-28T14:51:18.016419",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "data = np.array(data) #put pandas dataframe into numpy array\n",
    "m, n = data.shape #m is amount of rows, n is amount of features + 1\n",
    "np.random.shuffle(data) #randomly rearrange training examples\n",
    "\n",
    "# Split data into training and validation datasets\n",
    "\n",
    "data_validate = data[0:1000].T # validation set\n",
    "Y_validate = data_validate[0] # labels\n",
    "X_validate = data_validate[1:n] # features\n",
    "X_validate = X_validate / 255 # let pixel brightness be a scale from 0-1 rather than 0-255\n",
    "\n",
    "data_train = data[1000:m].T # training set\n",
    "Y_train = data_train[0] # labels\n",
    "X_train = data_train[1:n] #features\n",
    "X_train = X_train / 255 # let pixel brightness be a scale from 0-1 rather than 0-255\n",
    "_,m_train = X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c246ae96",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:18.442648Z",
     "iopub.status.busy": "2023-06-28T14:51:18.442296Z",
     "iopub.status.idle": "2023-06-28T14:51:18.448640Z",
     "shell.execute_reply": "2023-06-28T14:51:18.447592Z"
    },
    "papermill": {
     "duration": 0.0126,
     "end_time": "2023-06-28T14:51:18.450439",
     "exception": false,
     "start_time": "2023-06-28T14:51:18.437839",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(784,)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train[:, 0].shape # should output how many pixels in each image\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d030c225",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:18.459640Z",
     "iopub.status.busy": "2023-06-28T14:51:18.459294Z",
     "iopub.status.idle": "2023-06-28T14:51:18.465090Z",
     "shell.execute_reply": "2023-06-28T14:51:18.464233Z"
    },
    "papermill": {
     "duration": 0.012671,
     "end_time": "2023-06-28T14:51:18.466913",
     "exception": false,
     "start_time": "2023-06-28T14:51:18.454242",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3, 0, 4, ..., 8, 9, 5])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_train #the classification (0-9) for each training example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6a26ce20",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:18.475537Z",
     "iopub.status.busy": "2023-06-28T14:51:18.475241Z",
     "iopub.status.idle": "2023-06-28T14:51:18.485660Z",
     "shell.execute_reply": "2023-06-28T14:51:18.484741Z"
    },
    "papermill": {
     "duration": 0.016862,
     "end_time": "2023-06-28T14:51:18.487458",
     "exception": false,
     "start_time": "2023-06-28T14:51:18.470596",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def init_params():\n",
    "    W1 = np.random.rand(10, 784) - 0.5\n",
    "    b1 = np.random.rand(10, 1) - 0.5\n",
    "    W2 = np.random.rand(10, 10) - 0.5\n",
    "    b2 = np.random.rand(10, 1) - 0.5\n",
    "    return W1, b1, W2, b2\n",
    "\n",
    "#define ReLU (recitified linear unit) activation function\n",
    "def ReLU(Z):\n",
    "    return np.maximum(0, Z)\n",
    "\n",
    "#define softmax output layer function\n",
    "def softmax(Z):\n",
    "    output_layer = np.exp(Z) / sum(np.exp(Z))\n",
    "    return output_layer # returns output layer of probabilties (0-1)\n",
    "    \n",
    "def forward_prop(W1, b1, W2, b2, X):\n",
    "    Z1 = W1.dot(X) + b1 # weighted sum of input layer\n",
    "    A1 = ReLU(Z1) # use ReLU activation function on weighted sum of input layer to get first hidden layer\n",
    "    Z2 = W2.dot(A1) + b2 # weighted sum of first hidden layer layer\n",
    "    A2 = softmax(Z2) # use softmax activation function on weighted sum of first hidden layer to get output layer of probabilities\n",
    "    return Z1, A1, Z2, A2\n",
    "\n",
    "def ReLU_deriv(Z):\n",
    "    return Z > 0 #if Z > 0, derivative will be derivative of x (1 or TRUE), if Z <= 0, derivative will be derivative of 0 (0 or FALSE)\n",
    "\n",
    "def one_hot(Y):\n",
    "    one_hot_Y = np.zeros((Y.size, Y.max() + 1)) # Creates matrix of zeros, with dimensions m (number of examples) and n (possible output classes)\n",
    "    one_hot_Y[np.arange(Y.size), Y] = 1\n",
    "    one_hot_Y = one_hot_Y.T\n",
    "    return one_hot_Y\n",
    "\n",
    "def back_prop(Z1, A1, Z2, A2, W1, W2, X, Y):\n",
    "    one_hot_Y = one_hot(Y)\n",
    "    dZ2 = A2 - one_hot_Y\n",
    "    dW2 = 1 / m * dZ2.dot(A1.T)\n",
    "    db2 = 1 / m * np.sum(dZ2)\n",
    "    dZ1 = W2.T.dot(dZ2) * ReLU_deriv(Z1)\n",
    "    dW1 = 1 / m * dZ1.dot(X.T)\n",
    "    db1 = 1 / m * np.sum(dZ1)\n",
    "    return dW1, db1, dW2, db2\n",
    "\n",
    "def update_params(W1, b1, W2, b2, dW1, db1, dW2, db2, alpha):\n",
    "    W1 = W1 - alpha * dW1\n",
    "    b1 = b1 - alpha * db1    \n",
    "    W2 = W2 - alpha * dW2  \n",
    "    b2 = b2 - alpha * db2    \n",
    "    return W1, b1, W2, b2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "8f7a7ed7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:18.496555Z",
     "iopub.status.busy": "2023-06-28T14:51:18.496240Z",
     "iopub.status.idle": "2023-06-28T14:51:18.503268Z",
     "shell.execute_reply": "2023-06-28T14:51:18.502293Z"
    },
    "papermill": {
     "duration": 0.013685,
     "end_time": "2023-06-28T14:51:18.505015",
     "exception": false,
     "start_time": "2023-06-28T14:51:18.491330",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def get_predictions(A2):\n",
    "    return np.argmax(A2, 0)\n",
    "\n",
    "def get_accuracy(predictions, Y):\n",
    "    #print(predictions, Y)\n",
    "    return np.sum(predictions == Y) / Y.size\n",
    "    \n",
    "def gradient_descent(X, Y, iterations, alpha):\n",
    "    W1, b1, W2, b2 = init_params()\n",
    "    for i in range(iterations):\n",
    "        Z1, A1, Z2, A2 = forward_prop(W1, b1, W2, b2, X)\n",
    "        dW1, db1, dW2, db2 = back_prop(Z1, A1, Z2, A2, W1, W2, X, Y)\n",
    "        W1, b1, W2, b2 = update_params(W1, b1, W2, b2, dW1, db1, dW2, db2, alpha)\n",
    "        if i % 10 == 0:\n",
    "            print(\"Iteration: \", i)\n",
    "            print(\"Accuracy: \", get_accuracy(get_predictions(A2), Y))\n",
    "    return W1, b1, W2, b2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2d11510c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:51:18.515278Z",
     "iopub.status.busy": "2023-06-28T14:51:18.514997Z",
     "iopub.status.idle": "2023-06-28T14:56:56.004152Z",
     "shell.execute_reply": "2023-06-28T14:56:56.003436Z"
    },
    "papermill": {
     "duration": 337.496229,
     "end_time": "2023-06-28T14:56:56.006544",
     "exception": false,
     "start_time": "2023-06-28T14:51:18.510315",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration:  0\n",
      "Accuracy:  0.065\n",
      "Iteration:  10\n",
      "Accuracy:  0.3338536585365854\n",
      "Iteration:  20\n",
      "Accuracy:  0.4608536585365854\n",
      "Iteration:  30\n",
      "Accuracy:  0.5496341463414635\n",
      "Iteration:  40\n",
      "Accuracy:  0.6123658536585366\n",
      "Iteration:  50\n",
      "Accuracy:  0.6570487804878049\n",
      "Iteration:  60\n",
      "Accuracy:  0.6878048780487804\n",
      "Iteration:  70\n",
      "Accuracy:  0.7116829268292683\n",
      "Iteration:  80\n",
      "Accuracy:  0.7301951219512195\n",
      "Iteration:  90\n",
      "Accuracy:  0.7066585365853658\n",
      "Iteration:  100\n",
      "Accuracy:  0.7305853658536585\n",
      "Iteration:  110\n",
      "Accuracy:  0.7518048780487805\n",
      "Iteration:  120\n",
      "Accuracy:  0.7651463414634146\n",
      "Iteration:  130\n",
      "Accuracy:  0.7765365853658537\n",
      "Iteration:  140\n",
      "Accuracy:  0.7861707317073171\n",
      "Iteration:  150\n",
      "Accuracy:  0.7945853658536586\n",
      "Iteration:  160\n",
      "Accuracy:  0.8020975609756098\n",
      "Iteration:  170\n",
      "Accuracy:  0.8091951219512196\n",
      "Iteration:  180\n",
      "Accuracy:  0.8156585365853658\n",
      "Iteration:  190\n",
      "Accuracy:  0.8214146341463414\n",
      "Iteration:  200\n",
      "Accuracy:  0.8262682926829268\n",
      "Iteration:  210\n",
      "Accuracy:  0.8299512195121951\n",
      "Iteration:  220\n",
      "Accuracy:  0.8343170731707317\n",
      "Iteration:  230\n",
      "Accuracy:  0.8390243902439024\n",
      "Iteration:  240\n",
      "Accuracy:  0.8427560975609756\n",
      "Iteration:  250\n",
      "Accuracy:  0.8468292682926829\n",
      "Iteration:  260\n",
      "Accuracy:  0.8494390243902439\n",
      "Iteration:  270\n",
      "Accuracy:  0.8526585365853658\n",
      "Iteration:  280\n",
      "Accuracy:  0.8555121951219512\n",
      "Iteration:  290\n",
      "Accuracy:  0.8575365853658536\n",
      "Iteration:  300\n",
      "Accuracy:  0.8598780487804878\n",
      "Iteration:  310\n",
      "Accuracy:  0.8615121951219512\n",
      "Iteration:  320\n",
      "Accuracy:  0.8631951219512195\n",
      "Iteration:  330\n",
      "Accuracy:  0.8652926829268293\n",
      "Iteration:  340\n",
      "Accuracy:  0.8666585365853658\n",
      "Iteration:  350\n",
      "Accuracy:  0.8680243902439024\n",
      "Iteration:  360\n",
      "Accuracy:  0.8697560975609756\n",
      "Iteration:  370\n",
      "Accuracy:  0.8709756097560976\n",
      "Iteration:  380\n",
      "Accuracy:  0.8721463414634146\n",
      "Iteration:  390\n",
      "Accuracy:  0.8730487804878049\n",
      "Iteration:  400\n",
      "Accuracy:  0.8744878048780488\n",
      "Iteration:  410\n",
      "Accuracy:  0.8755121951219512\n",
      "Iteration:  420\n",
      "Accuracy:  0.876780487804878\n",
      "Iteration:  430\n",
      "Accuracy:  0.8776585365853659\n",
      "Iteration:  440\n",
      "Accuracy:  0.878609756097561\n",
      "Iteration:  450\n",
      "Accuracy:  0.8793658536585366\n",
      "Iteration:  460\n",
      "Accuracy:  0.8803414634146342\n",
      "Iteration:  470\n",
      "Accuracy:  0.8811951219512195\n",
      "Iteration:  480\n",
      "Accuracy:  0.882\n",
      "Iteration:  490\n",
      "Accuracy:  0.8827073170731707\n",
      "Iteration:  500\n",
      "Accuracy:  0.8831951219512195\n",
      "Iteration:  510\n",
      "Accuracy:  0.8837073170731707\n",
      "Iteration:  520\n",
      "Accuracy:  0.8844390243902439\n",
      "Iteration:  530\n",
      "Accuracy:  0.8851463414634146\n",
      "Iteration:  540\n",
      "Accuracy:  0.8861707317073171\n",
      "Iteration:  550\n",
      "Accuracy:  0.8868536585365854\n",
      "Iteration:  560\n",
      "Accuracy:  0.8873170731707317\n",
      "Iteration:  570\n",
      "Accuracy:  0.8878536585365854\n",
      "Iteration:  580\n",
      "Accuracy:  0.8885853658536585\n",
      "Iteration:  590\n",
      "Accuracy:  0.8893414634146342\n",
      "Iteration:  600\n",
      "Accuracy:  0.8896829268292683\n",
      "Iteration:  610\n",
      "Accuracy:  0.8903414634146342\n",
      "Iteration:  620\n",
      "Accuracy:  0.8908292682926829\n",
      "Iteration:  630\n",
      "Accuracy:  0.8915121951219512\n",
      "Iteration:  640\n",
      "Accuracy:  0.8919268292682927\n",
      "Iteration:  650\n",
      "Accuracy:  0.8925609756097561\n",
      "Iteration:  660\n",
      "Accuracy:  0.8930243902439025\n",
      "Iteration:  670\n",
      "Accuracy:  0.8934390243902439\n",
      "Iteration:  680\n",
      "Accuracy:  0.894\n",
      "Iteration:  690\n",
      "Accuracy:  0.8943658536585366\n",
      "Iteration:  700\n",
      "Accuracy:  0.894609756097561\n",
      "Iteration:  710\n",
      "Accuracy:  0.8951707317073171\n",
      "Iteration:  720\n",
      "Accuracy:  0.895609756097561\n",
      "Iteration:  730\n",
      "Accuracy:  0.8959268292682927\n",
      "Iteration:  740\n",
      "Accuracy:  0.8964390243902439\n",
      "Iteration:  750\n",
      "Accuracy:  0.8967560975609756\n",
      "Iteration:  760\n",
      "Accuracy:  0.8969268292682927\n",
      "Iteration:  770\n",
      "Accuracy:  0.8971951219512195\n",
      "Iteration:  780\n",
      "Accuracy:  0.8974634146341464\n",
      "Iteration:  790\n",
      "Accuracy:  0.897829268292683\n",
      "Iteration:  800\n",
      "Accuracy:  0.8979756097560976\n",
      "Iteration:  810\n",
      "Accuracy:  0.8981951219512195\n",
      "Iteration:  820\n",
      "Accuracy:  0.8985121951219512\n",
      "Iteration:  830\n",
      "Accuracy:  0.8988536585365854\n",
      "Iteration:  840\n",
      "Accuracy:  0.8992439024390244\n",
      "Iteration:  850\n",
      "Accuracy:  0.8993414634146342\n",
      "Iteration:  860\n",
      "Accuracy:  0.8995365853658537\n",
      "Iteration:  870\n",
      "Accuracy:  0.8998780487804878\n",
      "Iteration:  880\n",
      "Accuracy:  0.9001707317073171\n",
      "Iteration:  890\n",
      "Accuracy:  0.9005853658536586\n",
      "Iteration:  900\n",
      "Accuracy:  0.9007317073170732\n",
      "Iteration:  910\n",
      "Accuracy:  0.9011707317073171\n",
      "Iteration:  920\n",
      "Accuracy:  0.901390243902439\n",
      "Iteration:  930\n",
      "Accuracy:  0.9017804878048781\n",
      "Iteration:  940\n",
      "Accuracy:  0.9021951219512195\n",
      "Iteration:  950\n",
      "Accuracy:  0.9025121951219512\n",
      "Iteration:  960\n",
      "Accuracy:  0.9028536585365854\n",
      "Iteration:  970\n",
      "Accuracy:  0.9030975609756098\n",
      "Iteration:  980\n",
      "Accuracy:  0.9033658536585366\n",
      "Iteration:  990\n",
      "Accuracy:  0.9037073170731708\n",
      "Iteration:  1000\n",
      "Accuracy:  0.9039512195121951\n",
      "Iteration:  1010\n",
      "Accuracy:  0.9044390243902439\n",
      "Iteration:  1020\n",
      "Accuracy:  0.9047804878048781\n",
      "Iteration:  1030\n",
      "Accuracy:  0.9049268292682927\n",
      "Iteration:  1040\n",
      "Accuracy:  0.9050243902439025\n",
      "Iteration:  1050\n",
      "Accuracy:  0.9052439024390244\n",
      "Iteration:  1060\n",
      "Accuracy:  0.9055121951219512\n",
      "Iteration:  1070\n",
      "Accuracy:  0.9057073170731708\n",
      "Iteration:  1080\n",
      "Accuracy:  0.9059756097560976\n",
      "Iteration:  1090\n",
      "Accuracy:  0.9062926829268293\n",
      "Iteration:  1100\n",
      "Accuracy:  0.9067317073170732\n",
      "Iteration:  1110\n",
      "Accuracy:  0.9069024390243903\n",
      "Iteration:  1120\n",
      "Accuracy:  0.9070975609756098\n",
      "Iteration:  1130\n",
      "Accuracy:  0.9074390243902439\n",
      "Iteration:  1140\n",
      "Accuracy:  0.9077560975609756\n",
      "Iteration:  1150\n",
      "Accuracy:  0.9078536585365854\n",
      "Iteration:  1160\n",
      "Accuracy:  0.9080243902439025\n",
      "Iteration:  1170\n",
      "Accuracy:  0.9081951219512195\n",
      "Iteration:  1180\n",
      "Accuracy:  0.9083414634146342\n",
      "Iteration:  1190\n",
      "Accuracy:  0.9085121951219512\n",
      "Iteration:  1200\n",
      "Accuracy:  0.9086585365853659\n",
      "Iteration:  1210\n",
      "Accuracy:  0.9089268292682927\n",
      "Iteration:  1220\n",
      "Accuracy:  0.9092439024390244\n",
      "Iteration:  1230\n",
      "Accuracy:  0.9095365853658537\n",
      "Iteration:  1240\n",
      "Accuracy:  0.9096829268292683\n",
      "Iteration:  1250\n",
      "Accuracy:  0.91\n",
      "Iteration:  1260\n",
      "Accuracy:  0.9103414634146342\n",
      "Iteration:  1270\n",
      "Accuracy:  0.9106341463414634\n",
      "Iteration:  1280\n",
      "Accuracy:  0.9106829268292683\n",
      "Iteration:  1290\n",
      "Accuracy:  0.910829268292683\n",
      "Iteration:  1300\n",
      "Accuracy:  0.9111219512195122\n",
      "Iteration:  1310\n",
      "Accuracy:  0.9112682926829269\n",
      "Iteration:  1320\n",
      "Accuracy:  0.9114390243902439\n",
      "Iteration:  1330\n",
      "Accuracy:  0.9116585365853659\n",
      "Iteration:  1340\n",
      "Accuracy:  0.9119024390243903\n",
      "Iteration:  1350\n",
      "Accuracy:  0.9120731707317074\n",
      "Iteration:  1360\n",
      "Accuracy:  0.9121707317073171\n",
      "Iteration:  1370\n",
      "Accuracy:  0.9124878048780488\n",
      "Iteration:  1380\n",
      "Accuracy:  0.9127317073170732\n",
      "Iteration:  1390\n",
      "Accuracy:  0.913\n",
      "Iteration:  1400\n",
      "Accuracy:  0.9130487804878049\n",
      "Iteration:  1410\n",
      "Accuracy:  0.9132682926829269\n",
      "Iteration:  1420\n",
      "Accuracy:  0.9133902439024391\n",
      "Iteration:  1430\n",
      "Accuracy:  0.913609756097561\n",
      "Iteration:  1440\n",
      "Accuracy:  0.9138536585365854\n",
      "Iteration:  1450\n",
      "Accuracy:  0.9140975609756098\n",
      "Iteration:  1460\n",
      "Accuracy:  0.9141707317073171\n",
      "Iteration:  1470\n",
      "Accuracy:  0.9143170731707317\n",
      "Iteration:  1480\n",
      "Accuracy:  0.9144878048780488\n",
      "Iteration:  1490\n",
      "Accuracy:  0.9145365853658537\n",
      "Iteration:  1500\n",
      "Accuracy:  0.9146585365853659\n",
      "Iteration:  1510\n",
      "Accuracy:  0.914829268292683\n",
      "Iteration:  1520\n",
      "Accuracy:  0.9149024390243903\n",
      "Iteration:  1530\n",
      "Accuracy:  0.9150243902439025\n",
      "Iteration:  1540\n",
      "Accuracy:  0.9151219512195122\n",
      "Iteration:  1550\n",
      "Accuracy:  0.9152439024390244\n",
      "Iteration:  1560\n",
      "Accuracy:  0.9154878048780488\n",
      "Iteration:  1570\n",
      "Accuracy:  0.9155609756097561\n",
      "Iteration:  1580\n",
      "Accuracy:  0.9157560975609756\n",
      "Iteration:  1590\n",
      "Accuracy:  0.9159268292682927\n",
      "Iteration:  1600\n",
      "Accuracy:  0.9161219512195122\n",
      "Iteration:  1610\n",
      "Accuracy:  0.916219512195122\n",
      "Iteration:  1620\n",
      "Accuracy:  0.9163414634146342\n",
      "Iteration:  1630\n",
      "Accuracy:  0.9165121951219513\n",
      "Iteration:  1640\n",
      "Accuracy:  0.916609756097561\n",
      "Iteration:  1650\n",
      "Accuracy:  0.9167804878048781\n",
      "Iteration:  1660\n",
      "Accuracy:  0.9168780487804878\n",
      "Iteration:  1670\n",
      "Accuracy:  0.9169024390243903\n",
      "Iteration:  1680\n",
      "Accuracy:  0.9170243902439025\n",
      "Iteration:  1690\n",
      "Accuracy:  0.9170243902439025\n",
      "Iteration:  1700\n",
      "Accuracy:  0.9172682926829269\n",
      "Iteration:  1710\n",
      "Accuracy:  0.9174634146341464\n",
      "Iteration:  1720\n",
      "Accuracy:  0.9176341463414635\n",
      "Iteration:  1730\n",
      "Accuracy:  0.9176585365853659\n",
      "Iteration:  1740\n",
      "Accuracy:  0.9177317073170732\n",
      "Iteration:  1750\n",
      "Accuracy:  0.9179512195121952\n",
      "Iteration:  1760\n",
      "Accuracy:  0.9179756097560976\n",
      "Iteration:  1770\n",
      "Accuracy:  0.918219512195122\n",
      "Iteration:  1780\n",
      "Accuracy:  0.918219512195122\n",
      "Iteration:  1790\n",
      "Accuracy:  0.9183902439024391\n",
      "Iteration:  1800\n",
      "Accuracy:  0.918609756097561\n",
      "Iteration:  1810\n",
      "Accuracy:  0.9187560975609756\n",
      "Iteration:  1820\n",
      "Accuracy:  0.9187317073170732\n",
      "Iteration:  1830\n",
      "Accuracy:  0.918829268292683\n",
      "Iteration:  1840\n",
      "Accuracy:  0.9188048780487805\n",
      "Iteration:  1850\n",
      "Accuracy:  0.9190243902439025\n",
      "Iteration:  1860\n",
      "Accuracy:  0.9190487804878049\n",
      "Iteration:  1870\n",
      "Accuracy:  0.9192682926829269\n",
      "Iteration:  1880\n",
      "Accuracy:  0.9193902439024391\n",
      "Iteration:  1890\n",
      "Accuracy:  0.9194146341463415\n",
      "Iteration:  1900\n",
      "Accuracy:  0.9194878048780488\n",
      "Iteration:  1910\n",
      "Accuracy:  0.9195609756097561\n",
      "Iteration:  1920\n",
      "Accuracy:  0.9196341463414635\n",
      "Iteration:  1930\n",
      "Accuracy:  0.9198048780487805\n",
      "Iteration:  1940\n",
      "Accuracy:  0.9199512195121952\n",
      "Iteration:  1950\n",
      "Accuracy:  0.9201951219512196\n",
      "Iteration:  1960\n",
      "Accuracy:  0.9201707317073171\n",
      "Iteration:  1970\n",
      "Accuracy:  0.9202682926829269\n",
      "Iteration:  1980\n",
      "Accuracy:  0.9203414634146342\n",
      "Iteration:  1990\n",
      "Accuracy:  0.9204878048780488\n",
      "Iteration:  2000\n",
      "Accuracy:  0.9205853658536586\n",
      "Iteration:  2010\n",
      "Accuracy:  0.9206829268292683\n",
      "Iteration:  2020\n",
      "Accuracy:  0.9207073170731708\n",
      "Iteration:  2030\n",
      "Accuracy:  0.9207804878048781\n",
      "Iteration:  2040\n",
      "Accuracy:  0.9208536585365854\n",
      "Iteration:  2050\n",
      "Accuracy:  0.9210731707317074\n",
      "Iteration:  2060\n",
      "Accuracy:  0.9210731707317074\n",
      "Iteration:  2070\n",
      "Accuracy:  0.9210975609756098\n",
      "Iteration:  2080\n",
      "Accuracy:  0.9211951219512196\n",
      "Iteration:  2090\n",
      "Accuracy:  0.9212682926829269\n",
      "Iteration:  2100\n",
      "Accuracy:  0.9214146341463415\n",
      "Iteration:  2110\n",
      "Accuracy:  0.9215121951219513\n",
      "Iteration:  2120\n",
      "Accuracy:  0.9214878048780488\n",
      "Iteration:  2130\n",
      "Accuracy:  0.921609756097561\n",
      "Iteration:  2140\n",
      "Accuracy:  0.9218048780487805\n",
      "Iteration:  2150\n",
      "Accuracy:  0.922\n",
      "Iteration:  2160\n",
      "Accuracy:  0.9221219512195122\n",
      "Iteration:  2170\n",
      "Accuracy:  0.9221707317073171\n",
      "Iteration:  2180\n",
      "Accuracy:  0.922219512195122\n",
      "Iteration:  2190\n",
      "Accuracy:  0.9222682926829269\n",
      "Iteration:  2200\n",
      "Accuracy:  0.9225121951219513\n",
      "Iteration:  2210\n",
      "Accuracy:  0.9225609756097561\n",
      "Iteration:  2220\n",
      "Accuracy:  0.922609756097561\n",
      "Iteration:  2230\n",
      "Accuracy:  0.9226829268292683\n",
      "Iteration:  2240\n",
      "Accuracy:  0.9227560975609757\n",
      "Iteration:  2250\n",
      "Accuracy:  0.9229268292682927\n",
      "Iteration:  2260\n",
      "Accuracy:  0.923\n",
      "Iteration:  2270\n",
      "Accuracy:  0.9230487804878049\n",
      "Iteration:  2280\n",
      "Accuracy:  0.9231219512195122\n",
      "Iteration:  2290\n",
      "Accuracy:  0.9231463414634147\n",
      "Iteration:  2300\n",
      "Accuracy:  0.9233658536585366\n",
      "Iteration:  2310\n",
      "Accuracy:  0.9234634146341464\n",
      "Iteration:  2320\n",
      "Accuracy:  0.9235853658536586\n",
      "Iteration:  2330\n",
      "Accuracy:  0.9236341463414635\n",
      "Iteration:  2340\n",
      "Accuracy:  0.9237560975609757\n",
      "Iteration:  2350\n",
      "Accuracy:  0.9239024390243903\n",
      "Iteration:  2360\n",
      "Accuracy:  0.9240487804878049\n",
      "Iteration:  2370\n",
      "Accuracy:  0.9241219512195122\n",
      "Iteration:  2380\n",
      "Accuracy:  0.9242682926829269\n",
      "Iteration:  2390\n",
      "Accuracy:  0.9243902439024391\n",
      "Iteration:  2400\n",
      "Accuracy:  0.9243658536585366\n",
      "Iteration:  2410\n",
      "Accuracy:  0.9244878048780488\n",
      "Iteration:  2420\n",
      "Accuracy:  0.9246341463414635\n",
      "Iteration:  2430\n",
      "Accuracy:  0.9247317073170732\n",
      "Iteration:  2440\n",
      "Accuracy:  0.9248048780487805\n",
      "Iteration:  2450\n",
      "Accuracy:  0.9250731707317074\n",
      "Iteration:  2460\n",
      "Accuracy:  0.9252439024390244\n",
      "Iteration:  2470\n",
      "Accuracy:  0.9252682926829269\n",
      "Iteration:  2480\n",
      "Accuracy:  0.9254146341463415\n",
      "Iteration:  2490\n",
      "Accuracy:  0.9254634146341464\n",
      "Iteration:  2500\n",
      "Accuracy:  0.9255121951219513\n",
      "Iteration:  2510\n",
      "Accuracy:  0.9256341463414635\n",
      "Iteration:  2520\n",
      "Accuracy:  0.9256829268292683\n",
      "Iteration:  2530\n",
      "Accuracy:  0.9258536585365854\n",
      "Iteration:  2540\n",
      "Accuracy:  0.9259756097560976\n",
      "Iteration:  2550\n",
      "Accuracy:  0.9260731707317074\n",
      "Iteration:  2560\n",
      "Accuracy:  0.9262682926829269\n",
      "Iteration:  2570\n",
      "Accuracy:  0.9264146341463415\n",
      "Iteration:  2580\n",
      "Accuracy:  0.926439024390244\n",
      "Iteration:  2590\n",
      "Accuracy:  0.9265609756097561\n",
      "Iteration:  2600\n",
      "Accuracy:  0.9267317073170732\n",
      "Iteration:  2610\n",
      "Accuracy:  0.9268048780487805\n",
      "Iteration:  2620\n",
      "Accuracy:  0.9269756097560976\n",
      "Iteration:  2630\n",
      "Accuracy:  0.9271219512195122\n",
      "Iteration:  2640\n",
      "Accuracy:  0.9271707317073171\n",
      "Iteration:  2650\n",
      "Accuracy:  0.9272439024390244\n",
      "Iteration:  2660\n",
      "Accuracy:  0.9273902439024391\n",
      "Iteration:  2670\n",
      "Accuracy:  0.9274146341463415\n",
      "Iteration:  2680\n",
      "Accuracy:  0.9275365853658537\n",
      "Iteration:  2690\n",
      "Accuracy:  0.9276341463414635\n",
      "Iteration:  2700\n",
      "Accuracy:  0.9276585365853659\n",
      "Iteration:  2710\n",
      "Accuracy:  0.9277317073170732\n",
      "Iteration:  2720\n",
      "Accuracy:  0.9277804878048781\n",
      "Iteration:  2730\n",
      "Accuracy:  0.9279268292682927\n",
      "Iteration:  2740\n",
      "Accuracy:  0.9280487804878049\n",
      "Iteration:  2750\n",
      "Accuracy:  0.9281463414634147\n",
      "Iteration:  2760\n",
      "Accuracy:  0.9281219512195122\n",
      "Iteration:  2770\n",
      "Accuracy:  0.9281463414634147\n",
      "Iteration:  2780\n",
      "Accuracy:  0.9283170731707318\n",
      "Iteration:  2790\n",
      "Accuracy:  0.9284878048780488\n",
      "Iteration:  2800\n",
      "Accuracy:  0.9286341463414635\n",
      "Iteration:  2810\n",
      "Accuracy:  0.9287317073170732\n",
      "Iteration:  2820\n",
      "Accuracy:  0.9288780487804879\n",
      "Iteration:  2830\n",
      "Accuracy:  0.929\n",
      "Iteration:  2840\n",
      "Accuracy:  0.9290487804878049\n",
      "Iteration:  2850\n",
      "Accuracy:  0.9291951219512196\n",
      "Iteration:  2860\n",
      "Accuracy:  0.9291707317073171\n",
      "Iteration:  2870\n",
      "Accuracy:  0.929219512195122\n",
      "Iteration:  2880\n",
      "Accuracy:  0.9293658536585366\n",
      "Iteration:  2890\n",
      "Accuracy:  0.9294878048780488\n",
      "Iteration:  2900\n",
      "Accuracy:  0.9295121951219513\n",
      "Iteration:  2910\n",
      "Accuracy:  0.9295853658536586\n",
      "Iteration:  2920\n",
      "Accuracy:  0.9296829268292683\n",
      "Iteration:  2930\n",
      "Accuracy:  0.9298048780487805\n",
      "Iteration:  2940\n",
      "Accuracy:  0.9298780487804879\n",
      "Iteration:  2950\n",
      "Accuracy:  0.9298780487804879\n",
      "Iteration:  2960\n",
      "Accuracy:  0.9298048780487805\n",
      "Iteration:  2970\n",
      "Accuracy:  0.929829268292683\n",
      "Iteration:  2980\n",
      "Accuracy:  0.9299512195121952\n",
      "Iteration:  2990\n",
      "Accuracy:  0.93\n",
      "Iteration:  3000\n",
      "Accuracy:  0.9301463414634147\n",
      "Iteration:  3010\n",
      "Accuracy:  0.9301463414634147\n",
      "Iteration:  3020\n",
      "Accuracy:  0.9301707317073171\n",
      "Iteration:  3030\n",
      "Accuracy:  0.9301951219512196\n",
      "Iteration:  3040\n",
      "Accuracy:  0.9303414634146342\n",
      "Iteration:  3050\n",
      "Accuracy:  0.930439024390244\n",
      "Iteration:  3060\n",
      "Accuracy:  0.9304634146341464\n",
      "Iteration:  3070\n",
      "Accuracy:  0.9304878048780488\n",
      "Iteration:  3080\n",
      "Accuracy:  0.9305121951219513\n",
      "Iteration:  3090\n",
      "Accuracy:  0.9305853658536586\n",
      "Iteration:  3100\n",
      "Accuracy:  0.9306829268292683\n",
      "Iteration:  3110\n",
      "Accuracy:  0.9308048780487805\n",
      "Iteration:  3120\n",
      "Accuracy:  0.9308780487804879\n",
      "Iteration:  3130\n",
      "Accuracy:  0.9309268292682927\n",
      "Iteration:  3140\n",
      "Accuracy:  0.9309512195121952\n",
      "Iteration:  3150\n",
      "Accuracy:  0.9310975609756098\n",
      "Iteration:  3160\n",
      "Accuracy:  0.9312926829268293\n",
      "Iteration:  3170\n",
      "Accuracy:  0.9313414634146342\n",
      "Iteration:  3180\n",
      "Accuracy:  0.9313414634146342\n",
      "Iteration:  3190\n",
      "Accuracy:  0.931439024390244\n",
      "Iteration:  3200\n",
      "Accuracy:  0.9316341463414635\n",
      "Iteration:  3210\n",
      "Accuracy:  0.9317073170731708\n",
      "Iteration:  3220\n",
      "Accuracy:  0.9317317073170732\n",
      "Iteration:  3230\n",
      "Accuracy:  0.9317560975609757\n",
      "Iteration:  3240\n",
      "Accuracy:  0.9318536585365854\n",
      "Iteration:  3250\n",
      "Accuracy:  0.9319024390243903\n",
      "Iteration:  3260\n",
      "Accuracy:  0.9320243902439025\n",
      "Iteration:  3270\n",
      "Accuracy:  0.9321219512195122\n",
      "Iteration:  3280\n",
      "Accuracy:  0.9320975609756098\n",
      "Iteration:  3290\n",
      "Accuracy:  0.932219512195122\n",
      "Iteration:  3300\n",
      "Accuracy:  0.9323170731707318\n",
      "Iteration:  3310\n",
      "Accuracy:  0.9323902439024391\n",
      "Iteration:  3320\n",
      "Accuracy:  0.9324878048780488\n",
      "Iteration:  3330\n",
      "Accuracy:  0.9325609756097561\n",
      "Iteration:  3340\n",
      "Accuracy:  0.9326585365853659\n",
      "Iteration:  3350\n",
      "Accuracy:  0.9327073170731708\n",
      "Iteration:  3360\n",
      "Accuracy:  0.9327317073170732\n",
      "Iteration:  3370\n",
      "Accuracy:  0.9328048780487805\n",
      "Iteration:  3380\n",
      "Accuracy:  0.9329512195121952\n",
      "Iteration:  3390\n",
      "Accuracy:  0.9329756097560976\n",
      "Iteration:  3400\n",
      "Accuracy:  0.9329756097560976\n",
      "Iteration:  3410\n",
      "Accuracy:  0.9330243902439025\n",
      "Iteration:  3420\n",
      "Accuracy:  0.9330975609756098\n",
      "Iteration:  3430\n",
      "Accuracy:  0.9331951219512196\n",
      "Iteration:  3440\n",
      "Accuracy:  0.9331707317073171\n",
      "Iteration:  3450\n",
      "Accuracy:  0.9332682926829269\n",
      "Iteration:  3460\n",
      "Accuracy:  0.9332682926829269\n",
      "Iteration:  3470\n",
      "Accuracy:  0.9332439024390244\n",
      "Iteration:  3480\n",
      "Accuracy:  0.9333170731707318\n",
      "Iteration:  3490\n",
      "Accuracy:  0.9335365853658537\n",
      "Iteration:  3500\n",
      "Accuracy:  0.9335365853658537\n",
      "Iteration:  3510\n",
      "Accuracy:  0.9335853658536586\n",
      "Iteration:  3520\n",
      "Accuracy:  0.9335609756097561\n",
      "Iteration:  3530\n",
      "Accuracy:  0.9337073170731708\n",
      "Iteration:  3540\n",
      "Accuracy:  0.9337560975609757\n",
      "Iteration:  3550\n",
      "Accuracy:  0.9338048780487805\n",
      "Iteration:  3560\n",
      "Accuracy:  0.9339512195121952\n",
      "Iteration:  3570\n",
      "Accuracy:  0.9340243902439025\n",
      "Iteration:  3580\n",
      "Accuracy:  0.9341707317073171\n",
      "Iteration:  3590\n",
      "Accuracy:  0.9341951219512196\n",
      "Iteration:  3600\n",
      "Accuracy:  0.9342682926829269\n",
      "Iteration:  3610\n",
      "Accuracy:  0.9343170731707318\n",
      "Iteration:  3620\n",
      "Accuracy:  0.9343414634146342\n",
      "Iteration:  3630\n",
      "Accuracy:  0.9343414634146342\n",
      "Iteration:  3640\n",
      "Accuracy:  0.9344146341463415\n",
      "Iteration:  3650\n",
      "Accuracy:  0.9344634146341464\n",
      "Iteration:  3660\n",
      "Accuracy:  0.934609756097561\n",
      "Iteration:  3670\n",
      "Accuracy:  0.934609756097561\n",
      "Iteration:  3680\n",
      "Accuracy:  0.934609756097561\n",
      "Iteration:  3690\n",
      "Accuracy:  0.9346585365853659\n",
      "Iteration:  3700\n",
      "Accuracy:  0.9346829268292683\n",
      "Iteration:  3710\n",
      "Accuracy:  0.9347317073170732\n",
      "Iteration:  3720\n",
      "Accuracy:  0.9349024390243903\n",
      "Iteration:  3730\n",
      "Accuracy:  0.9349756097560976\n",
      "Iteration:  3740\n",
      "Accuracy:  0.9350487804878049\n",
      "Iteration:  3750\n",
      "Accuracy:  0.9351219512195122\n",
      "Iteration:  3760\n",
      "Accuracy:  0.9351707317073171\n",
      "Iteration:  3770\n",
      "Accuracy:  0.9351707317073171\n",
      "Iteration:  3780\n",
      "Accuracy:  0.9352926829268293\n",
      "Iteration:  3790\n",
      "Accuracy:  0.9353658536585366\n",
      "Iteration:  3800\n",
      "Accuracy:  0.9353902439024391\n",
      "Iteration:  3810\n",
      "Accuracy:  0.9353658536585366\n",
      "Iteration:  3820\n",
      "Accuracy:  0.9353902439024391\n",
      "Iteration:  3830\n",
      "Accuracy:  0.9354146341463415\n",
      "Iteration:  3840\n",
      "Accuracy:  0.9354634146341464\n",
      "Iteration:  3850\n",
      "Accuracy:  0.9354634146341464\n",
      "Iteration:  3860\n",
      "Accuracy:  0.9354878048780488\n",
      "Iteration:  3870\n",
      "Accuracy:  0.9355609756097562\n",
      "Iteration:  3880\n",
      "Accuracy:  0.9356829268292683\n",
      "Iteration:  3890\n",
      "Accuracy:  0.935829268292683\n",
      "Iteration:  3900\n",
      "Accuracy:  0.9358536585365854\n",
      "Iteration:  3910\n",
      "Accuracy:  0.9358780487804879\n",
      "Iteration:  3920\n",
      "Accuracy:  0.9359268292682927\n",
      "Iteration:  3930\n",
      "Accuracy:  0.936\n",
      "Iteration:  3940\n",
      "Accuracy:  0.936\n",
      "Iteration:  3950\n",
      "Accuracy:  0.9360731707317074\n",
      "Iteration:  3960\n",
      "Accuracy:  0.9360731707317074\n",
      "Iteration:  3970\n",
      "Accuracy:  0.9361707317073171\n",
      "Iteration:  3980\n",
      "Accuracy:  0.9362682926829269\n",
      "Iteration:  3990\n",
      "Accuracy:  0.9362926829268293\n",
      "Iteration:  4000\n",
      "Accuracy:  0.9362926829268293\n",
      "Iteration:  4010\n",
      "Accuracy:  0.9363902439024391\n",
      "Iteration:  4020\n",
      "Accuracy:  0.9364146341463415\n",
      "Iteration:  4030\n",
      "Accuracy:  0.936439024390244\n",
      "Iteration:  4040\n",
      "Accuracy:  0.9364634146341464\n",
      "Iteration:  4050\n",
      "Accuracy:  0.9365121951219513\n",
      "Iteration:  4060\n",
      "Accuracy:  0.9365121951219513\n",
      "Iteration:  4070\n",
      "Accuracy:  0.9365365853658537\n",
      "Iteration:  4080\n",
      "Accuracy:  0.9365365853658537\n",
      "Iteration:  4090\n",
      "Accuracy:  0.9366585365853659\n",
      "Iteration:  4100\n",
      "Accuracy:  0.9366829268292683\n",
      "Iteration:  4110\n",
      "Accuracy:  0.9367073170731708\n",
      "Iteration:  4120\n",
      "Accuracy:  0.9367804878048781\n",
      "Iteration:  4130\n",
      "Accuracy:  0.936829268292683\n",
      "Iteration:  4140\n",
      "Accuracy:  0.9369024390243903\n",
      "Iteration:  4150\n",
      "Accuracy:  0.9369024390243903\n",
      "Iteration:  4160\n",
      "Accuracy:  0.9369024390243903\n",
      "Iteration:  4170\n",
      "Accuracy:  0.9369512195121952\n",
      "Iteration:  4180\n",
      "Accuracy:  0.9370487804878049\n",
      "Iteration:  4190\n",
      "Accuracy:  0.9370243902439025\n",
      "Iteration:  4200\n",
      "Accuracy:  0.9370487804878049\n",
      "Iteration:  4210\n",
      "Accuracy:  0.9371707317073171\n",
      "Iteration:  4220\n",
      "Accuracy:  0.9372439024390244\n",
      "Iteration:  4230\n",
      "Accuracy:  0.9372439024390244\n",
      "Iteration:  4240\n",
      "Accuracy:  0.9373414634146342\n",
      "Iteration:  4250\n",
      "Accuracy:  0.9373170731707318\n",
      "Iteration:  4260\n",
      "Accuracy:  0.9374146341463415\n",
      "Iteration:  4270\n",
      "Accuracy:  0.9375121951219512\n",
      "Iteration:  4280\n",
      "Accuracy:  0.9375853658536585\n",
      "Iteration:  4290\n",
      "Accuracy:  0.9377073170731707\n",
      "Iteration:  4300\n",
      "Accuracy:  0.9377560975609756\n",
      "Iteration:  4310\n",
      "Accuracy:  0.937780487804878\n",
      "Iteration:  4320\n",
      "Accuracy:  0.9378780487804877\n",
      "Iteration:  4330\n",
      "Accuracy:  0.9379512195121951\n",
      "Iteration:  4340\n",
      "Accuracy:  0.9379756097560975\n",
      "Iteration:  4350\n",
      "Accuracy:  0.9379268292682926\n",
      "Iteration:  4360\n",
      "Accuracy:  0.938\n",
      "Iteration:  4370\n",
      "Accuracy:  0.9381951219512195\n",
      "Iteration:  4380\n",
      "Accuracy:  0.9382926829268292\n",
      "Iteration:  4390\n",
      "Accuracy:  0.9383658536585365\n",
      "Iteration:  4400\n",
      "Accuracy:  0.9383658536585365\n",
      "Iteration:  4410\n",
      "Accuracy:  0.9384390243902438\n",
      "Iteration:  4420\n",
      "Accuracy:  0.9384634146341463\n",
      "Iteration:  4430\n",
      "Accuracy:  0.9384390243902438\n",
      "Iteration:  4440\n",
      "Accuracy:  0.9384390243902438\n",
      "Iteration:  4450\n",
      "Accuracy:  0.9384390243902438\n",
      "Iteration:  4460\n",
      "Accuracy:  0.9384390243902438\n",
      "Iteration:  4470\n",
      "Accuracy:  0.9385121951219512\n",
      "Iteration:  4480\n",
      "Accuracy:  0.9384390243902438\n",
      "Iteration:  4490\n",
      "Accuracy:  0.9384878048780487\n",
      "Iteration:  4500\n",
      "Accuracy:  0.938560975609756\n",
      "Iteration:  4510\n",
      "Accuracy:  0.9386097560975609\n",
      "Iteration:  4520\n",
      "Accuracy:  0.9387073170731707\n",
      "Iteration:  4530\n",
      "Accuracy:  0.9387560975609756\n",
      "Iteration:  4540\n",
      "Accuracy:  0.9388048780487804\n",
      "Iteration:  4550\n",
      "Accuracy:  0.9388048780487804\n",
      "Iteration:  4560\n",
      "Accuracy:  0.9388536585365853\n",
      "Iteration:  4570\n",
      "Accuracy:  0.9389024390243902\n",
      "Iteration:  4580\n",
      "Accuracy:  0.9389268292682926\n",
      "Iteration:  4590\n",
      "Accuracy:  0.9389024390243902\n",
      "Iteration:  4600\n",
      "Accuracy:  0.9388780487804878\n",
      "Iteration:  4610\n",
      "Accuracy:  0.9389512195121951\n",
      "Iteration:  4620\n",
      "Accuracy:  0.9389512195121951\n",
      "Iteration:  4630\n",
      "Accuracy:  0.9390243902439024\n",
      "Iteration:  4640\n",
      "Accuracy:  0.939\n",
      "Iteration:  4650\n",
      "Accuracy:  0.939\n",
      "Iteration:  4660\n",
      "Accuracy:  0.9390243902439024\n",
      "Iteration:  4670\n",
      "Accuracy:  0.9390487804878048\n",
      "Iteration:  4680\n",
      "Accuracy:  0.9391219512195121\n",
      "Iteration:  4690\n",
      "Accuracy:  0.9392439024390243\n",
      "Iteration:  4700\n",
      "Accuracy:  0.9392682926829268\n",
      "Iteration:  4710\n",
      "Accuracy:  0.9392682926829268\n",
      "Iteration:  4720\n",
      "Accuracy:  0.9392682926829268\n",
      "Iteration:  4730\n",
      "Accuracy:  0.9393414634146341\n",
      "Iteration:  4740\n",
      "Accuracy:  0.9394146341463414\n",
      "Iteration:  4750\n",
      "Accuracy:  0.9394390243902438\n",
      "Iteration:  4760\n",
      "Accuracy:  0.9394878048780487\n",
      "Iteration:  4770\n",
      "Accuracy:  0.9395121951219512\n",
      "Iteration:  4780\n",
      "Accuracy:  0.9395365853658536\n",
      "Iteration:  4790\n",
      "Accuracy:  0.9396097560975609\n",
      "Iteration:  4800\n",
      "Accuracy:  0.9397560975609756\n",
      "Iteration:  4810\n",
      "Accuracy:  0.9398536585365853\n",
      "Iteration:  4820\n",
      "Accuracy:  0.9399268292682926\n",
      "Iteration:  4830\n",
      "Accuracy:  0.9399512195121951\n",
      "Iteration:  4840\n",
      "Accuracy:  0.9399268292682926\n",
      "Iteration:  4850\n",
      "Accuracy:  0.94\n",
      "Iteration:  4860\n",
      "Accuracy:  0.9400975609756097\n",
      "Iteration:  4870\n",
      "Accuracy:  0.940170731707317\n",
      "Iteration:  4880\n",
      "Accuracy:  0.9401951219512195\n",
      "Iteration:  4890\n",
      "Accuracy:  0.9402682926829268\n",
      "Iteration:  4900\n",
      "Accuracy:  0.9403170731707317\n",
      "Iteration:  4910\n",
      "Accuracy:  0.9403658536585365\n",
      "Iteration:  4920\n",
      "Accuracy:  0.9404146341463414\n",
      "Iteration:  4930\n",
      "Accuracy:  0.9404878048780487\n",
      "Iteration:  4940\n",
      "Accuracy:  0.9404634146341463\n",
      "Iteration:  4950\n",
      "Accuracy:  0.9405365853658536\n",
      "Iteration:  4960\n",
      "Accuracy:  0.9406097560975609\n",
      "Iteration:  4970\n",
      "Accuracy:  0.9406829268292682\n",
      "Iteration:  4980\n",
      "Accuracy:  0.9407560975609756\n",
      "Iteration:  4990\n",
      "Accuracy:  0.9408536585365853\n"
     ]
    }
   ],
   "source": [
    "W1, b1, W2, b2 = gradient_descent(X_train, Y_train, 5000, 0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f004ff7",
   "metadata": {
    "papermill": {
     "duration": 0.037997,
     "end_time": "2023-06-28T14:56:56.097939",
     "exception": false,
     "start_time": "2023-06-28T14:56:56.059942",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Now Let's See how the model does on the validation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a3ebbcc4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:56:56.168202Z",
     "iopub.status.busy": "2023-06-28T14:56:56.167851Z",
     "iopub.status.idle": "2023-06-28T14:56:56.174346Z",
     "shell.execute_reply": "2023-06-28T14:56:56.173675Z"
    },
    "papermill": {
     "duration": 0.040878,
     "end_time": "2023-06-28T14:56:56.176443",
     "exception": false,
     "start_time": "2023-06-28T14:56:56.135565",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def make_predictions(X, W1, b1, W2, b2):\n",
    "    _, _, _, A2 = forward_prop(W1, b1, W2, b2, X)\n",
    "    predictions = get_predictions(A2)\n",
    "    return predictions\n",
    "\n",
    "def test_prediction(index, W1, b1, W2, b2):\n",
    "    current_image = X_validate[:, index, None]\n",
    "    prediction = make_predictions(X_validate[:, index, None], W1, b1, W2, b2)\n",
    "    label = Y_validate[index]\n",
    "    print(\"Prediction: \", prediction)\n",
    "    print(\"Label: \", label)\n",
    "    \n",
    "    current_image = current_image.reshape((28, 28)) * 255\n",
    "    plt.gray()\n",
    "    plt.imshow(current_image, interpolation='nearest')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "58be91cb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:56:56.242832Z",
     "iopub.status.busy": "2023-06-28T14:56:56.242279Z",
     "iopub.status.idle": "2023-06-28T14:56:56.251246Z",
     "shell.execute_reply": "2023-06-28T14:56:56.250431Z"
    },
    "papermill": {
     "duration": 0.043529,
     "end_time": "2023-06-28T14:56:56.253077",
     "exception": false,
     "start_time": "2023-06-28T14:56:56.209548",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  94.39999999999999 %\n"
     ]
    }
   ],
   "source": [
    "validate_predictions = make_predictions(X_validate, W1, b1, W2, b2)\n",
    "print(\"Accuracy: \", 100 * get_accuracy(validate_predictions, Y_validate), \"%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7d5747d",
   "metadata": {
    "papermill": {
     "duration": 0.039945,
     "end_time": "2023-06-28T14:56:56.333466",
     "exception": false,
     "start_time": "2023-06-28T14:56:56.293521",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Let's look at some examples of how it does on the validation set:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "94e5dcb6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-28T14:56:56.405107Z",
     "iopub.status.busy": "2023-06-28T14:56:56.404798Z",
     "iopub.status.idle": "2023-06-28T14:56:59.582545Z",
     "shell.execute_reply": "2023-06-28T14:56:59.581532Z"
    },
    "papermill": {
     "duration": 3.212635,
     "end_time": "2023-06-28T14:56:59.584527",
     "exception": false,
     "start_time": "2023-06-28T14:56:56.371892",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [5]\n",
      "Label:  5\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [7]\n",
      "Label:  7\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [6]\n",
      "Label:  6\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [9]\n",
      "Label:  9\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [9]\n",
      "Label:  9\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [4]\n",
      "Label:  4\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [1]\n",
      "Label:  1\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [4]\n",
      "Label:  4\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [7]\n",
      "Label:  7\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [2]\n",
      "Label:  2\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [8]\n",
      "Label:  8\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [6]\n",
      "Label:  6\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [5]\n",
      "Label:  5\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [8]\n",
      "Label:  8\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [6]\n",
      "Label:  6\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [5]\n",
      "Label:  5\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [6]\n",
      "Label:  6\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaEAAAGdCAYAAAC7EMwUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAZ1UlEQVR4nO3df2hV9/3H8df1163NkjtCTO7NjCGIztGIo2qjWf1VZmq2uaot2BZK/EfaGQWJRaahmG7MFKFSWFbHSrF2q6t/TK1Qqc2miY7MToOiuOLSGestGoLB3RujvU79fP8Q77e3SdOc67155ybPB3zAe855e975eJKXn9x7z/U555wAADAwyroBAMDIRQgBAMwQQgAAM4QQAMAMIQQAMEMIAQDMEEIAADOEEADAzBjrBr7u7t27unz5srKzs+Xz+azbAQB45JxTd3e3CgsLNWpU/2udIRdCly9fVlFRkXUbAIAHFA6HNXHixH6PGXK/jsvOzrZuAQCQAgP5eZ62EHrzzTdVUlKihx56SDNnztSxY8cGVMev4ABgeBjIz/O0hNCePXu0fv161dbW6tSpU5o3b54qKyt16dKldJwOAJChfOm4i3ZZWZkeffRR7dixI77tBz/4gZYtW6b6+vp+a6PRqAKBQKpbAgAMskgkopycnH6PSflK6NatW2ptbVVFRUXC9oqKCrW0tPQ6PhaLKRqNJgwAwMiQ8hC6evWq7ty5o4KCgoTtBQUF6ujo6HV8fX29AoFAfPDKOAAYOdL2woSvPyHlnOvzSapNmzYpEonERzgcTldLAIAhJuXvE8rLy9Po0aN7rXo6Ozt7rY4kye/3y+/3p7oNAEAGSPlKaNy4cZo5c6YaGxsTtjc2Nqq8vDzVpwMAZLC03DGhpqZGL7zwgmbNmqW5c+fqD3/4gy5duqSXXnopHacDAGSotITQypUr1dXVpV/96le6cuWKSktLdfDgQRUXF6fjdACADJWW9wk9CN4nBADDg8n7hAAAGChCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABm0nIXbSBTvfPOO55rXnjhBc81LS0tnmvmzZvnuQYY6lgJAQDMEEIAADOEEADADCEEADBDCAEAzBBCAAAzhBAAwAwhBAAwQwgBAMwQQgAAM4QQAMAMIQQAMEMIAQDMcBdt4Cvmzp3rucY5NyjnWb58ueeaffv2ea4BBhMrIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGZ8Lpm7L6ZRNBpVIBCwbgMj1Pnz5z3XTJ48OQ2d9NbT0+O5hu8lWIpEIsrJyen3GFZCAAAzhBAAwAwhBAAwQwgBAMwQQgAAM4QQAMAMIQQAMEMIAQDMEEIAADOEEADADCEEADBDCAEAzIyxbgBIhzlz5iRVFwwGU9wJgP6wEgIAmCGEAABmUh5CdXV18vl8CYNfcQAA+pKW54QeeeQR/fWvf40/Hj16dDpOAwDIcGkJoTFjxrD6AQB8q7Q8J9TW1qbCwkKVlJTo2Wef1YULF77x2Fgspmg0mjAAACNDykOorKxM7777rg4dOqS33npLHR0dKi8vV1dXV5/H19fXKxAIxEdRUVGqWwIADFE+55xL5wl6eno0efJkbdy4UTU1Nb32x2IxxWKx+ONoNEoQ4YEl+z6hQ4cOea7JyspK6lxe9fT0eK4JBAJp6AQYmEgkopycnH6PSfubVbOysjR9+nS1tbX1ud/v98vv96e7DQDAEJT29wnFYjF9+umnCoVC6T4VACDDpDyEXn75ZTU3N6u9vV2ffPKJnnnmGUWjUVVVVaX6VACADJfyX8d98cUXeu6553T16lVNmDBBc+bM0fHjx1VcXJzqUwEAMlzKQ+j9999P9V8JeLZhw4ak6gbrRQbJ2L59u3ULQMpx7zgAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABm0v7Jql5Fo1E+DRIJkvmU1GQ+IVUa2jcwHTMm7Z9BCaTUQD5ZlZUQAMAMIQQAMEMIAQDMEEIAADOEEADADCEEADBDCAEAzBBCAAAzhBAAwAwhBAAwQwgBAMwQQgAAM4QQAMAMt+XFkDd37lzPNUP5btgA/h8rIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGa4gSmGvGXLllm3kHI7duywbgEYElgJAQDMEEIAADOEEADADCEEADBDCAEAzBBCAAAzhBAAwAwhBAAwQwgBAMwQQgAAM4QQAMAMIQQAMMMNTDGoqqurPdc89thjaejE1t/+9jfrFoAhgZUQAMAMIQQAMOM5hI4ePaqlS5eqsLBQPp9P+/fvT9jvnFNdXZ0KCws1fvx4LVy4UOfOnUtVvwCAYcRzCPX09GjGjBlqaGjoc/+2bdu0fft2NTQ06MSJEwoGg1q8eLG6u7sfuFkAwPDi+YUJlZWVqqys7HOfc05vvPGGamtrtWLFCknSrl27VFBQoN27d+vFF198sG4BAMNKSp8Tam9vV0dHhyoqKuLb/H6/FixYoJaWlj5rYrGYotFowgAAjAwpDaGOjg5JUkFBQcL2goKC+L6vq6+vVyAQiI+ioqJUtgQAGMLS8uo4n8+X8Ng512vbfZs2bVIkEomPcDicjpYAAENQSt+sGgwGJd1bEYVCofj2zs7OXquj+/x+v/x+fyrbAABkiJSuhEpKShQMBtXY2BjfduvWLTU3N6u8vDyVpwIADAOeV0LXr1/XZ599Fn/c3t6u06dPKzc3V5MmTdL69eu1detWTZkyRVOmTNHWrVv18MMP6/nnn09p4wCAzOc5hE6ePKlFixbFH9fU1EiSqqqq9M4772jjxo26efOm1qxZo2vXrqmsrEwff/yxsrOzU9c1AGBY8DnnnHUTXxWNRhUIBKzbwADcfw7QiyNHjniumTJliueawbRjxw7PNc8884znmqysLM81g/ntvW/fPs81v/71rz3X/Oc///FcAxuRSEQ5OTn9HsO94wAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZriLNpSXl5dUXXNzs+eaadOmJXUuSKNGef8/4927d9PQSeqEw2HPNT/+8Y8913z1M9AweLiLNgBgSCOEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGBmjHUDsPfd7343qbrvf//7nmuG2P1yM0oyNyMd6vM9ceJEzzU7duzwXPOTn/zEc40k/e9//0uqDgPHSggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZbmAKdXZ2JlV38uRJzzWzZs1K6lxIzp07d5Kqa21t9VzzwQcfeK6pra31XLNo0SLPNRs3bvRcI0m/+c1vkqrDwLESAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYbmELRaDSpun/+85+ea7iBafJOnz7tuebAgQNJnevVV19Nqs6r+fPne6558sknPdesXr3ac40kvf32255rOjo6kjrXSMVKCABghhACAJjxHEJHjx7V0qVLVVhYKJ/Pp/379yfsX7VqlXw+X8KYM2dOqvoFAAwjnkOop6dHM2bMUENDwzces2TJEl25ciU+Dh48+EBNAgCGJ88vTKisrFRlZWW/x/j9fgWDwaSbAgCMDGl5TqipqUn5+fmaOnWqVq9e3e/HR8diMUWj0YQBABgZUh5ClZWVeu+993T48GG9/vrrOnHihJ544gnFYrE+j6+vr1cgEIiPoqKiVLcEABiiUv4+oZUrV8b/XFpaqlmzZqm4uFgffvihVqxY0ev4TZs2qaamJv44Go0SRAAwQqT9zaqhUEjFxcVqa2vrc7/f75ff7093GwCAISjt7xPq6upSOBxWKBRK96kAABnG80ro+vXr+uyzz+KP29vbdfr0aeXm5io3N1d1dXV6+umnFQqFdPHiRW3evFl5eXlavnx5ShsHAGQ+zyF08uRJLVq0KP74/vM5VVVV2rFjh86ePat3331X//3vfxUKhbRo0SLt2bNH2dnZqesaADAs+JxzzrqJr4pGowoEAtZtYAB++MMfeq45efJk6hvJQOFw2HNNSUlJGjrp2/jx4z3X/PSnP/Vc09+b3r9JXl6e55pkTZs2zXPNV39TNNJFIhHl5OT0ewz3jgMAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmEn7J6ti+ErmbsGffPKJ55qysjLPNYPp888/91zz5JNPeq4ZM8b7t+vixYs910jS008/7blm1apVSZ0LIxsrIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGa4gSmSdv36dc81mzdv9lyzZ88ezzV5eXmea5KVlZXluSaZr2n06NGea0pLSz3XDEdffPFFUnXJXOPwhpUQAMAMIQQAMEMIAQDMEEIAADOEEADADCEEADBDCAEAzBBCAAAzhBAAwAwhBAAwQwgBAMwQQgAAM9zAFIOqubnZc8358+c91wzmDUyTOVcyNT6fz3ONc85zzVCXzM1If/7znyd1ro6OjqTqMHCshAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJjhBqYY8n772996rvnRj36Uhk6QagcOHPBcc+jQIc81Z86c8VyDwcFKCABghhACAJjxFEL19fWaPXu2srOzlZ+fr2XLlvX6rBfnnOrq6lRYWKjx48dr4cKFOnfuXEqbBgAMD55CqLm5WdXV1Tp+/LgaGxt1+/ZtVVRUqKenJ37Mtm3btH37djU0NOjEiRMKBoNavHixuru7U948ACCzeXphwkcffZTweOfOncrPz1dra6vmz58v55zeeOMN1dbWasWKFZKkXbt2qaCgQLt379aLL76Yus4BABnvgZ4TikQikqTc3FxJUnt7uzo6OlRRURE/xu/3a8GCBWppaenz74jFYopGowkDADAyJB1CzjnV1NTo8ccfV2lpqaT//zz2goKChGMLCgq+8bPa6+vrFQgE4qOoqCjZlgAAGSbpEFq7dq3OnDmjP//5z732+Xy+hMfOuV7b7tu0aZMikUh8hMPhZFsCAGSYpN6sum7dOh04cEBHjx7VxIkT49uDwaCkeyuiUCgU397Z2dlrdXSf3++X3+9Ppg0AQIbztBJyzmnt2rXau3evDh8+rJKSkoT9JSUlCgaDamxsjG+7deuWmpubVV5enpqOAQDDhqeVUHV1tXbv3q0PPvhA2dnZ8ed5AoGAxo8fL5/Pp/Xr12vr1q2aMmWKpkyZoq1bt+rhhx/W888/n5YvAACQuTyF0I4dOyRJCxcuTNi+c+dOrVq1SpK0ceNG3bx5U2vWrNG1a9dUVlamjz/+WNnZ2SlpGAAwfPicc866ia+KRqMKBALWbWAIGTPG+1OXa9asSepcr7zyiueaP/7xj55rvvoG74HavHmz55rB/Pbet2+f55ra2lrPNf/+978918BGJBJRTk5Ov8dw7zgAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBnuog0ASAvuog0AGNIIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmPIVQfX29Zs+erezsbOXn52vZsmU6f/58wjGrVq2Sz+dLGHPmzElp0wCA4cFTCDU3N6u6ulrHjx9XY2Ojbt++rYqKCvX09CQct2TJEl25ciU+Dh48mNKmAQDDwxgvB3/00UcJj3fu3Kn8/Hy1trZq/vz58e1+v1/BYDA1HQIAhq0Hek4oEolIknJzcxO2NzU1KT8/X1OnTtXq1avV2dn5jX9HLBZTNBpNGACAkcHnnHPJFDrn9NRTT+natWs6duxYfPuePXv0ne98R8XFxWpvb9crr7yi27dvq7W1VX6/v9ffU1dXp1dffTX5rwAAMCRFIhHl5OT0f5BL0po1a1xxcbELh8P9Hnf58mU3duxY95e//KXP/V9++aWLRCLxEQ6HnSQGg8FgZPiIRCLfmiWenhO6b926dTpw4ICOHj2qiRMn9ntsKBRScXGx2tra+tzv9/v7XCEBAIY/TyHknNO6deu0b98+NTU1qaSk5Ftrurq6FA6HFQqFkm4SADA8eXphQnV1tf70pz9p9+7dys7OVkdHhzo6OnTz5k1J0vXr1/Xyyy/rH//4hy5evKimpiYtXbpUeXl5Wr58eVq+AABABvPyPJC+4fd+O3fudM45d+PGDVdRUeEmTJjgxo4d6yZNmuSqqqrcpUuXBnyOSCRi/ntMBoPBYDz4GMhzQkm/Oi5dotGoAoGAdRsAgAc0kFfHce84AIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABmCCEAgBlCCABghhACAJghhAAAZgghAICZIRdCzjnrFgAAKTCQn+dDLoS6u7utWwAApMBAfp773BBbety9e1eXL19Wdna2fD5fwr5oNKqioiKFw2Hl5OQYdWiPebiHebiHebiHebhnKMyDc07d3d0qLCzUqFH9r3XGDFJPAzZq1ChNnDix32NycnJG9EV2H/NwD/NwD/NwD/Nwj/U8BAKBAR035H4dBwAYOQghAICZjAohv9+vLVu2yO/3W7diinm4h3m4h3m4h3m4J9PmYci9MAEAMHJk1EoIADC8EEIAADOEEADADCEEADCTUSH05ptvqqSkRA899JBmzpypY8eOWbc0qOrq6uTz+RJGMBi0bivtjh49qqVLl6qwsFA+n0/79+9P2O+cU11dnQoLCzV+/HgtXLhQ586ds2k2jb5tHlatWtXr+pgzZ45Ns2lSX1+v2bNnKzs7W/n5+Vq2bJnOnz+fcMxIuB4GMg+Zcj1kTAjt2bNH69evV21trU6dOqV58+apsrJSly5dsm5tUD3yyCO6cuVKfJw9e9a6pbTr6enRjBkz1NDQ0Of+bdu2afv27WpoaNCJEycUDAa1ePHiYXcfwm+bB0lasmRJwvVx8ODBQeww/Zqbm1VdXa3jx4+rsbFRt2/fVkVFhXp6euLHjITrYSDzIGXI9eAyxGOPPeZeeumlhG3Tpk1zv/zlL406GnxbtmxxM2bMsG7DlCS3b9+++OO7d++6YDDoXnvttfi2L7/80gUCAff73//eoMPB8fV5cM65qqoq99RTT5n0Y6Wzs9NJcs3Nzc65kXs9fH0enMuc6yEjVkK3bt1Sa2urKioqErZXVFSopaXFqCsbbW1tKiwsVElJiZ599llduHDBuiVT7e3t6ujoSLg2/H6/FixYMOKuDUlqampSfn6+pk6dqtWrV6uzs9O6pbSKRCKSpNzcXEkj93r4+jzclwnXQ0aE0NWrV3Xnzh0VFBQkbC8oKFBHR4dRV4OvrKxM7777rg4dOqS33npLHR0dKi8vV1dXl3VrZu7/+4/0a0OSKisr9d577+nw4cN6/fXXdeLECT3xxBOKxWLWraWFc041NTV6/PHHVVpaKmlkXg99zYOUOdfDkLuLdn++/tEOzrle24azysrK+J+nT5+uuXPnavLkydq1a5dqamoMO7M30q8NSVq5cmX8z6WlpZo1a5aKi4v14YcfasWKFYadpcfatWt15swZ/f3vf++1byRdD980D5lyPWTESigvL0+jR4/u9T+Zzs7OXv/jGUmysrI0ffp0tbW1Wbdi5v6rA7k2eguFQiouLh6W18e6det04MABHTlyJOGjX0ba9fBN89CXoXo9ZEQIjRs3TjNnzlRjY2PC9sbGRpWXlxt1ZS8Wi+nTTz9VKBSybsVMSUmJgsFgwrVx69YtNTc3j+hrQ5K6uroUDoeH1fXhnNPatWu1d+9eHT58WCUlJQn7R8r18G3z0Jchez0YvijCk/fff9+NHTvWvf322+5f//qXW79+vcvKynIXL160bm3QbNiwwTU1NbkLFy6448ePu5/97GcuOzt72M9Bd3e3O3XqlDt16pST5LZv3+5OnTrlPv/8c+ecc6+99poLBAJu79697uzZs+65555zoVDIRaNR485Tq7956O7udhs2bHAtLS2uvb3dHTlyxM2dO9d973vfG1bz8Itf/MIFAgHX1NTkrly5Eh83btyIHzMSrodvm4dMuh4yJoScc+53v/udKy4uduPGjXOPPvpowssRR4KVK1e6UCjkxo4d6woLC92KFSvcuXPnrNtKuyNHjjhJvUZVVZVz7t7Lcrds2eKCwaDz+/1u/vz57uzZs7ZNp0F/83Djxg1XUVHhJkyY4MaOHesmTZrkqqqq3KVLl6zbTqm+vn5JbufOnfFjRsL18G3zkEnXAx/lAAAwkxHPCQEAhidCCABghhACAJghhAAAZgghAIAZQggAYIYQAgCYIYQAAGYIIQCAGUIIAGCGEAIAmCGEAABm/g+M4uZkR3dmDAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [7]\n",
      "Label:  7\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [5]\n",
      "Label:  5\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [4]\n",
      "Label:  4\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_prediction(0, W1, b1, W2, b2)\n",
    "test_prediction(1, W1, b1, W2, b2)\n",
    "test_prediction(2, W1, b1, W2, b2)\n",
    "test_prediction(3, W1, b1, W2, b2)\n",
    "test_prediction(4, W1, b1, W2, b2)\n",
    "test_prediction(5, W1, b1, W2, b2)\n",
    "test_prediction(6, W1, b1, W2, b2)\n",
    "test_prediction(7, W1, b1, W2, b2)\n",
    "test_prediction(8, W1, b1, W2, b2)\n",
    "test_prediction(9, W1, b1, W2, b2)\n",
    "test_prediction(10, W1, b1, W2, b2)\n",
    "test_prediction(11, W1, b1, W2, b2)\n",
    "test_prediction(12, W1, b1, W2, b2)\n",
    "test_prediction(13, W1, b1, W2, b2)\n",
    "test_prediction(14, W1, b1, W2, b2)\n",
    "test_prediction(15, W1, b1, W2, b2)\n",
    "test_prediction(16, W1, b1, W2, b2)\n",
    "test_prediction(17, W1, b1, W2, b2)\n",
    "test_prediction(18, W1, b1, W2, b2)\n",
    "test_prediction(19, W1, b1, W2, b2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d8cf1ce0",
   "metadata": {
    "papermill": {
     "duration": 0.035414,
     "end_time": "2023-06-28T14:56:59.656018",
     "exception": false,
     "start_time": "2023-06-28T14:56:59.620604",
     "status": "completed"
    },
    "tags": []
   },
   "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": "3.10.10"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 354.288434,
   "end_time": "2023-06-28T14:57:00.513636",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2023-06-28T14:51:06.225202",
   "version": "2.4.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}