neural.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gzip\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_mnist_images(filename):\n",
    "    with open(filename, 'rb') as f:\n",
    "        f.read(16)\n",
    "        buffer = f.read()\n",
    "        data = np.frombuffer(buffer, dtype=np.uint8)\n",
    "        data = data.reshape(-1, 28, 28)\n",
    "    return np.array(data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_mnist_labels(filename):\n",
    "    with open(filename, 'rb') as f:\n",
    "        f.read(8)\n",
    "        buffer = f.read()\n",
    "        labels = np.frombuffer(buffer, dtype=np.uint8)\n",
    "    return np.array(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train = load_mnist_images('data/train-images.idx3-ubyte')\n",
    "y_train = load_mnist_labels('data/train-labels.idx1-ubyte')\n",
    "x_test = load_mnist_images('data/t10k-images.idx3-ubyte')\n",
    "y_test = load_mnist_labels('data/t10k-labels.idx1-ubyte')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train_flat = x_train.reshape(x_train.shape[0], -1).T / 255.0\n",
    "x_test_flat = x_test.reshape(x_test.shape[0], -1).T / 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "m = x_train_flat.shape[1]\n",
    "shuffled_indices = np.random.permutation(m)\n",
    "x_train_flat = x_train_flat[:, shuffled_indices]\n",
    "y_train = y_train[shuffled_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_dev = x_train_flat[:, :1000]\n",
    "Y_dev = y_train[:1000]\n",
    "X_train = x_train_flat[:, 1000:]\n",
    "Y_train = y_train[1000:]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train shape: (784, 59000), Y_train shape: (59000,)\n",
      "X_dev shape: (784, 1000), Y_dev shape: (1000,)\n"
     ]
    }
   ],
   "source": [
    "print(f\"X_train shape: {X_train.shape}, Y_train shape: {Y_train.shape}\")\n",
    "print(f\"X_dev shape: {X_dev.shape}, Y_dev shape: {Y_dev.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def init_params():\n",
    "    W1 = np.random.rand(128, 784) - 0.5\n",
    "    b1 = np.random.rand(128, 1) - 0.5\n",
    "    W2 = np.random.rand(10, 128) - 0.5\n",
    "    b2 = np.random.rand(10, 1) - 0.5\n",
    "    return W1, b1, W2, b2\n",
    "\n",
    "def ReLU(Z):\n",
    "    return np.maximum(Z, 0)\n",
    "\n",
    "def softmax(Z):\n",
    "    expZ = np.exp(Z - np.max(Z, axis=0, keepdims=True))\n",
    "    return expZ / expZ.sum(axis=0, keepdims=True)\n",
    "\n",
    "def forward_prop(W1, b1, W2, b2, X):\n",
    "    Z1 = W1.dot(X) + b1\n",
    "    A1 = ReLU(Z1)\n",
    "    Z2 = W2.dot(A1) + b2\n",
    "    A2 = softmax(Z2)\n",
    "    return Z1, A1, Z2, A2\n",
    "\n",
    "def ReLU_deriv(Z):\n",
    "    return Z > 0\n",
    "\n",
    "def one_hot(Y):\n",
    "    one_hot_Y = np.zeros((Y.size, Y.max() + 1))\n",
    "    one_hot_Y[np.arange(Y.size), Y] = 1\n",
    "    return one_hot_Y.T\n",
    "\n",
    "def backward_prop(Z1, A1, Z2, A2, W1, W2, X, Y):\n",
    "    m = Y.size\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, axis=1, keepdims=True)\n",
    "    dZ1 = W2.T.dot(dZ2) * ReLU_deriv(Z1)\n",
    "    dW1 = 1 / m * dZ1.dot(X.T)\n",
    "    db1 = 1 / m * np.sum(dZ1, axis=1, keepdims=True)\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\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_predictions(A2):\n",
    "    return np.argmax(A2, 0)\n",
    "\n",
    "def get_accuracy(predictions, Y):\n",
    "    return np.sum(predictions == Y) / Y.size\n",
    "\n",
    "def gradient_descent(X, Y, alpha, iterations):\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 = backward_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(\"Iteração: \", i)\n",
    "            predictions = get_predictions(A2)\n",
    "            accuracy = get_accuracy(predictions, Y)\n",
    "            print(f\"Precisão: {accuracy:.2f}\")\n",
    "    return W1, b1, W2, b2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iteration:  0\n",
      "Accuracy: 0.09\n",
      "Iteration:  10\n",
      "Accuracy: 0.46\n",
      "Iteration:  20\n",
      "Accuracy: 0.62\n",
      "Iteration:  30\n",
      "Accuracy: 0.69\n",
      "Iteration:  40\n",
      "Accuracy: 0.73\n",
      "Iteration:  50\n",
      "Accuracy: 0.76\n",
      "Iteration:  60\n",
      "Accuracy: 0.78\n",
      "Iteration:  70\n",
      "Accuracy: 0.79\n",
      "Iteration:  80\n",
      "Accuracy: 0.81\n",
      "Iteration:  90\n",
      "Accuracy: 0.82\n",
      "Test set accuracy: 0.83\n"
     ]
    }
   ],
   "source": [
    "W1, b1, W2, b2 = gradient_descent(X_train, Y_train, 0.10, 100)\n",
    "\n",
    "Z1, A1, Z2, A2 = forward_prop(W1, b1, W2, b2, x_test_flat)\n",
    "predictions = get_predictions(A2)\n",
    "test_accuracy = get_accuracy(predictions, y_test)\n",
    "print(f\"Test set accuracy: {test_accuracy:.2f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "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_train[:, index, None]\n",
    "    prediction = make_predictions(X_train[:, index, None], W1, b1, W2, b2)\n",
    "    label = Y_train[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": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [8]\n",
      "Label:  8\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaAAAAGdCAYAAABU0qcqAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAcUklEQVR4nO3df2xV9f3H8dflR6+o7cVS29vKD8sPxQjUyaRWtEPpWrqF8WubOpOBIRBcMUPmj3SZou5HJy7OuTFclg10ExUSgUi2Rqy2RFcwVJF0m5V2VUpoyyTj3lJoQfr5/kG8X6+04Lnc23dbno/kk9BzzrvnzYeTvjj3nn6uzznnBABALxtk3QAA4MJEAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMDEEOsGvqirq0sHDx5UcnKyfD6fdTsAAI+cc2pra1NWVpYGDer5PqfPBdDBgwc1atQo6zYAAOepqalJI0eO7HF/n3sJLjk52boFAEAcnOvnecICaM2aNbryyit10UUXKTc3V++8886XquNlNwAYGM718zwhAfTyyy9r5cqVWrVqld59913l5OSoqKhIhw4dSsTpAAD9kUuAadOmuZKSksjXp06dcllZWa6srOyctaFQyEliMBgMRj8foVDorD/v434HdOLECdXU1KigoCCybdCgQSooKFB1dfUZx3d2diocDkcNAMDAF/cA+uSTT3Tq1CllZGREbc/IyFBLS8sZx5eVlSkQCEQGT8ABwIXB/Cm40tJShUKhyGhqarJuCQDQC+L+e0BpaWkaPHiwWltbo7a3trYqGAyecbzf75ff7493GwCAPi7ud0BJSUmaOnWqKioqItu6urpUUVGhvLy8eJ8OANBPJWQlhJUrV2rhwoX66le/qmnTpunpp59We3u77r777kScDgDQDyUkgG6//Xb997//1SOPPKKWlhZdd911Ki8vP+PBBADAhcvnnHPWTXxeOBxWIBCwbgMAcJ5CoZBSUlJ63G/+FBwA4MJEAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATBBAAAATBBAAwMQQ6wbQf1133XWeayZMmOC5prCw0HPN2LFjPddI0tSpUz3XbNq0KaZzefXBBx94rikvL4/pXP/85z9jqgO84A4IAGCCAAIAmIh7AD366KPy+XxRY+LEifE+DQCgn0vIe0DXXnutXn/99f8/yRDeagIAREtIMgwZMkTBYDAR3xoAMEAk5D2gffv2KSsrS2PHjtVdd92l/fv393hsZ2enwuFw1AAADHxxD6Dc3FytX79e5eXlWrt2rRobG3XLLbeora2t2+PLysoUCAQiY9SoUfFuCQDQB8U9gIqLi/Wd73xHU6ZMUVFRkf72t7/pyJEj2rhxY7fHl5aWKhQKRUZTU1O8WwIA9EEJfzpg+PDhuuqqq1RfX9/tfr/fL7/fn+g2AAB9TMJ/D+jo0aNqaGhQZmZmok8FAOhH4h5A999/v6qqqvTRRx/pH//4h+bNm6fBgwfrzjvvjPepAAD9WNxfgjtw4IDuvPNOHT58WJdffrluvvlm7dy5U5dffnm8TwUA6Md8zjln3cTnhcNhBQIB6zb6raFDh3quWbRoUUzn+t3vfue5Jpb+ELv29vaY6n7+8597rvnVr37luebTTz/1XIP+IxQKKSUlpcf9rAUHADBBAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADARMI/kA6xi2VR1k2bNnmuKSgo8FyD/uGSSy6Jqe4Xv/iF55rjx497rvnNb37juQYDB3dAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATBBAAAATPuecs27i88LhcEyrQA9ETz31lOeaFStWxL+RHvTW6scbNmzwXBNLb5J01VVXea758MMPPddMnz7dc838+fM913zrW9/yXBOrWOZ88uTJnmv+85//eK6BjVAopJSUlB73cwcEADBBAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABIuR9mF//OMfPdcsXrw4AZ1075NPPvFck56enoBOLgyDBnn//+Kf//znmM71/e9/P6Y6r5577jnPNXfffXcCOkEisBgpAKBPIoAAACYIIACACQIIAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJgggAAAJgggAIAJAggAYGKIdQMAvpyuri7PNUuXLo3pXNOnT/dcM27cOM81xcXFnmuGDRvmueb48eOea5B43AEBAEwQQAAAE54DaMeOHZo9e7aysrLk8/m0ZcuWqP3OOT3yyCPKzMzUsGHDVFBQoH379sWrXwDAAOE5gNrb25WTk6M1a9Z0u3/16tV65pln9Oyzz2rXrl265JJLVFRUpI6OjvNuFgAwcHh+CKG4uLjHNw6dc3r66af1k5/8RHPmzJEkPf/888rIyNCWLVt0xx13nF+3AIABI67vATU2NqqlpUUFBQWRbYFAQLm5uaquru62prOzU+FwOGoAAAa+uAZQS0uLJCkjIyNqe0ZGRmTfF5WVlSkQCETGqFGj4tkSAKCPMn8KrrS0VKFQKDKampqsWwIA9IK4BlAwGJQktba2Rm1vbW2N7Psiv9+vlJSUqAEAGPjiGkDZ2dkKBoOqqKiIbAuHw9q1a5fy8vLieSoAQD/n+Sm4o0ePqr6+PvJ1Y2Oj9uzZo9TUVI0ePVorVqzQz372M02YMEHZ2dl6+OGHlZWVpblz58azbwBAP+c5gHbv3q1bb7018vXKlSslSQsXLtT69ev14IMPqr29XUuXLtWRI0d08803q7y8XBdddFH8ugYA9HueA2jGjBlyzvW43+fz6fHHH9fjjz9+Xo0BOH8nTpyIqe7VV1/1XLNixQrPNbG853vNNdd4rnn33Xc91yDxzJ+CAwBcmAggAIAJAggAYIIAAgCYIIAAACYIIACACQIIAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJgggAAAJjyvho3e884773iuWbx4cQI66d6QId4vn4kTJ3qu+fznT31Zn376qeeaWH3961/3XFNUVOS5Zt68eZ5rRowY4blGOv1Jxb2hra3Nc82HH36YgE5ggTsgAIAJAggAYIIAAgCYIIAAACYIIACACQIIAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJgggAAAJnzOOWfdxOeFw2EFAgHrNvqE5ORkzzUbN270XBPLwpi96c033/Rc85e//CWmc02bNs1zzV133eW5JpZ/W5z297//3XPNggULYjpXR0dHTHU4LRQKKSUlpcf93AEBAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATBBAAAATBBAAwAQBBAAwwWKkA8z111/vuWb79u0xneuyyy6LqQ7obR9//HFMdbfeeqvnmo8++iimcw1ELEYKAOiTCCAAgAkCCABgggACAJgggAAAJgggAIAJAggAYIIAAgCYIIAAACYIIACACQIIAGCCAAIAmGAxUuimm26Kqe7JJ5/0XHPjjTd6rvH5fJ5r0PvKy8s918Ry7Z1tcct4+9///ue5ZtSoUZ5rjh075rmmP2AxUgBAn0QAAQBMeA6gHTt2aPbs2crKypLP59OWLVui9i9atEg+ny9qzJo1K179AgAGCM8B1N7erpycHK1Zs6bHY2bNmqXm5ubIePHFF8+rSQDAwDPEa0FxcbGKi4vPeozf71cwGIy5KQDAwJeQ94AqKyuVnp6uq6++Wvfcc48OHz7c47GdnZ0Kh8NRAwAw8MU9gGbNmqXnn39eFRUVeuKJJ1RVVaXi4mKdOnWq2+PLysoUCAQiI5ZHGAEA/Y/nl+DO5Y477oj8efLkyZoyZYrGjRunyspKzZw584zjS0tLtXLlysjX4XCYEAKAC0DCH8MeO3as0tLSVF9f3+1+v9+vlJSUqAEAGPgSHkAHDhzQ4cOHlZmZmehTAQD6Ec8vwR09ejTqbqaxsVF79uxRamqqUlNT9dhjj2nBggUKBoNqaGjQgw8+qPHjx6uoqCiujQMA+jfPAbR7927deuutka8/e/9m4cKFWrt2rfbu3avnnntOR44cUVZWlgoLC/XTn/5Ufr8/fl0DAPo9FiNFr9q8ebPnmjlz5iSgk+4dP37cc82BAwc812RlZXmuqamp8Vzz9ttve66RpNraWs81sfzCeU5OjueaJ554wnNNYWGh55pYvfLKK55rvv3tbyegE3ssRgoA6JMIIACACQIIAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJgggAAAJgggAIAJAggAYIIAAgCYIIAAACbi/pHcQH+2aNEizzWvvfaa55rLLrvMc81HH33kuaave//99z3X3H///Z5r3nrrLc81kmL6hOabbrrJc00sH9jZ3Nzsuaav4Q4IAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJgggAAAJgggAIAJAggAYIIAAgCYIIAAACYIIACACRYjBT4nloVFQ6FQr9TgtNraWs81a9eujelcDz30kOeaYDDouWb06NGea1iMFACAGBFAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADDBYqTA5wwbNsxzDQuL4vN8Pp91C/0Gd0AAABMEEADABAEEADBBAAEATBBAAAATBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMsBgpetVrr73muWbOnDkJ6KR7r7zyiuea2bNne645fPiw5xrE7sorr+y1cznneu1c/R13QAAAEwQQAMCEpwAqKyvTDTfcoOTkZKWnp2vu3Lmqq6uLOqajo0MlJSUaMWKELr30Ui1YsECtra1xbRoA0P95CqCqqiqVlJRo586d2r59u06ePKnCwkK1t7dHjrnvvvv06quvatOmTaqqqtLBgwc1f/78uDcOAOjfPD2EUF5eHvX1+vXrlZ6erpqaGuXn5ysUCulPf/qTNmzYoNtuu02StG7dOl1zzTXauXOnbrzxxvh1DgDo187rPaDPPoo4NTVVklRTU6OTJ0+qoKAgcszEiRM1evRoVVdXd/s9Ojs7FQ6HowYAYOCLOYC6urq0YsUKTZ8+XZMmTZIktbS0KCkpScOHD486NiMjQy0tLd1+n7KyMgUCgcgYNWpUrC0BAPqRmAOopKREtbW1eumll86rgdLSUoVCochoamo6r+8HAOgfYvpF1OXLl2vbtm3asWOHRo4cGdkeDAZ14sQJHTlyJOouqLW1VcFgsNvv5ff75ff7Y2kDANCPeboDcs5p+fLl2rx5s9544w1lZ2dH7Z86daqGDh2qioqKyLa6ujrt379feXl58ekYADAgeLoDKikp0YYNG7R161YlJydH3tcJBAIaNmyYAoGAFi9erJUrVyo1NVUpKSm69957lZeXxxNwAIAongJo7dq1kqQZM2ZEbV+3bp0WLVokSfr1r3+tQYMGacGCBers7FRRUZF+//vfx6VZAMDA4XN9bOW8cDisQCBg3QYSJCkpyXNNbW2t55rx48d7ronV7t27Pdds3LjRc82WLVs81/R1X/nKVzzXxPKL7d/97nc910iSz+fzXLNr1y7PNfn5+Z5rTp486bmmt4VCIaWkpPS4n7XgAAAmCCAAgAkCCABgggACAJgggAAAJgggAIAJAggAYIIAAgCYIIAAACYIIACACQIIAGCCAAIAmCCAAAAmWA0bfd6kSZM817z22msxnaunT+4FvqwJEyZ4rmloaEhAJ/ZYDRsA0CcRQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATBBAAAATBBAAwAQBBAAwMcS6AeBcamtrPdfcdtttMZ2roqLCc01mZmZM54Lk8/k818SyfvKJEyc810jSU0895bnm448/julcFyLugAAAJgggAIAJAggAYIIAAgCYIIAAACYIIACACQIIAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJjwuVhW9kugcDisQCBg3QYuUKNHj/Zcs2TJEs811157reeauXPneq7pTbEs5NrY2Oi5JpbFPrdt2+a5RpLef//9mOpwWigUUkpKSo/7uQMCAJgggAAAJgggAIAJAggAYIIAAgCYIIAAACYIIACACQIIAGCCAAIAmCCAAAAmCCAAgAkCCABggsVIAQAJwWKkAIA+iQACAJjwFEBlZWW64YYblJycrPT0dM2dO1d1dXVRx8yYMUM+ny9qLFu2LK5NAwD6P08BVFVVpZKSEu3cuVPbt2/XyZMnVVhYqPb29qjjlixZoubm5shYvXp1XJsGAPR/Q7wcXF5eHvX1+vXrlZ6erpqaGuXn50e2X3zxxQoGg/HpEAAwIJ3Xe0ChUEiSlJqaGrX9hRdeUFpamiZNmqTS0lIdO3asx+/R2dmpcDgcNQAAFwAXo1OnTrlvfvObbvr06VHb//CHP7jy8nK3d+9e99e//tVdccUVbt68eT1+n1WrVjlJDAaDwRhgIxQKnTVHYg6gZcuWuTFjxrimpqazHldRUeEkufr6+m73d3R0uFAoFBlNTU3mk8ZgMBiM8x/nCiBP7wF9Zvny5dq2bZt27NihkSNHnvXY3NxcSVJ9fb3GjRt3xn6/3y+/3x9LGwCAfsxTADnndO+992rz5s2qrKxUdnb2OWv27NkjScrMzIypQQDAwOQpgEpKSrRhwwZt3bpVycnJamlpkSQFAgENGzZMDQ0N2rBhg77xjW9oxIgR2rt3r+677z7l5+drypQpCfkLAAD6KS/v+6iH1/nWrVvnnHNu//79Lj8/36Wmpjq/3+/Gjx/vHnjggXO+Dvh5oVDI/HVLBoPBYJz/ONfPfhYjBQAkBIuRAgD6JAIIAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJgggAAAJgggAIAJAggAYIIAAgCYIIAAACYIIACACQIIAGCCAAIAmCCAAAAmCCAAgAkCCABgggACAJgggAAAJgggAIAJAggAYIIAAgCYIIAAACYIIACAiT4XQM456xYAAHFwrp/nfS6A2trarFsAAMTBuX6e+1wfu+Xo6urSwYMHlZycLJ/PF7UvHA5r1KhRampqUkpKilGH9piH05iH05iH05iH0/rCPDjn1NbWpqysLA0a1PN9zpBe7OlLGTRokEaOHHnWY1JSUi7oC+wzzMNpzMNpzMNpzMNp1vMQCATOeUyfewkOAHBhIIAAACb6VQD5/X6tWrVKfr/fuhVTzMNpzMNpzMNpzMNp/Wke+txDCACAC0O/ugMCAAwcBBAAwAQBBAAwQQABAEz0mwBas2aNrrzySl100UXKzc3VO++8Y91Sr3v00Ufl8/mixsSJE63bSrgdO3Zo9uzZysrKks/n05YtW6L2O+f0yCOPKDMzU8OGDVNBQYH27dtn02wCnWseFi1adMb1MWvWLJtmE6SsrEw33HCDkpOTlZ6errlz56quri7qmI6ODpWUlGjEiBG69NJLtWDBArW2thp1nBhfZh5mzJhxxvWwbNkyo4671y8C6OWXX9bKlSu1atUqvfvuu8rJyVFRUZEOHTpk3Vqvu/baa9Xc3BwZb731lnVLCdfe3q6cnBytWbOm2/2rV6/WM888o2effVa7du3SJZdcoqKiInV0dPRyp4l1rnmQpFmzZkVdHy+++GIvdph4VVVVKikp0c6dO7V9+3adPHlShYWFam9vjxxz33336dVXX9WmTZtUVVWlgwcPav78+YZdx9+XmQdJWrJkSdT1sHr1aqOOe+D6gWnTprmSkpLI16dOnXJZWVmurKzMsKvet2rVKpeTk2PdhilJbvPmzZGvu7q6XDAYdE8++WRk25EjR5zf73cvvviiQYe944vz4JxzCxcudHPmzDHpx8qhQ4ecJFdVVeWcO/1vP3ToULdp06bIMf/+97+dJFddXW3VZsJ9cR6cc+5rX/ua++EPf2jX1JfQ5++ATpw4oZqaGhUUFES2DRo0SAUFBaqurjbszMa+ffuUlZWlsWPH6q677tL+/futWzLV2NiolpaWqOsjEAgoNzf3grw+KisrlZ6erquvvlr33HOPDh8+bN1SQoVCIUlSamqqJKmmpkYnT56Muh4mTpyo0aNHD+jr4Yvz8JkXXnhBaWlpmjRpkkpLS3Xs2DGL9nrU5xYj/aJPPvlEp06dUkZGRtT2jIwMffDBB0Zd2cjNzdX69et19dVXq7m5WY899phuueUW1dbWKjk52bo9Ey0tLZLU7fXx2b4LxaxZszR//nxlZ2eroaFBP/7xj1VcXKzq6moNHjzYur246+rq0ooVKzR9+nRNmjRJ0unrISkpScOHD486diBfD93NgyR973vf05gxY5SVlaW9e/fqoYceUl1dnV555RXDbqP1+QDC/ysuLo78ecqUKcrNzdWYMWO0ceNGLV682LAz9AV33HFH5M+TJ0/WlClTNG7cOFVWVmrmzJmGnSVGSUmJamtrL4j3Qc+mp3lYunRp5M+TJ09WZmamZs6cqYaGBo0bN6632+xWn38JLi0tTYMHDz7jKZbW1lYFg0GjrvqG4cOH66qrrlJ9fb11K2Y+uwa4Ps40duxYpaWlDcjrY/ny5dq2bZvefPPNqI9vCQaDOnHihI4cORJ1/EC9Hnqah+7k5uZKUp+6Hvp8ACUlJWnq1KmqqKiIbOvq6lJFRYXy8vIMO7N39OhRNTQ0KDMz07oVM9nZ2QoGg1HXRzgc1q5duy746+PAgQM6fPjwgLo+nHNavny5Nm/erDfeeEPZ2dlR+6dOnaqhQ4dGXQ91dXXav3//gLoezjUP3dmzZ48k9a3rwfopiC/jpZdecn6/361fv97961//ckuXLnXDhw93LS0t1q31qh/96EeusrLSNTY2urffftsVFBS4tLQ0d+jQIevWEqqtrc2999577r333nOS3FNPPeXee+899/HHHzvnnPvlL3/phg8f7rZu3er27t3r5syZ47Kzs93x48eNO4+vs81DW1ubu//++111dbVrbGx0r7/+urv++uvdhAkTXEdHh3XrcXPPPfe4QCDgKisrXXNzc2QcO3YscsyyZcvc6NGj3RtvvOF2797t8vLyXF5enmHX8Xeueaivr3ePP/642717t2tsbHRbt251Y8eOdfn5+cadR+sXAeScc7/97W/d6NGjXVJSkps2bZrbuXOndUu97vbbb3eZmZkuKSnJXXHFFe7222939fX11m0l3JtvvukknTEWLlzonDv9KPbDDz/sMjIynN/vdzNnznR1dXW2TSfA2ebh2LFjrrCw0F1++eVu6NChbsyYMW7JkiUD7j9p3f39Jbl169ZFjjl+/Lj7wQ9+4C677DJ38cUXu3nz5rnm5ma7phPgXPOwf/9+l5+f71JTU53f73fjx493DzzwgAuFQraNfwEfxwAAMNHn3wMCAAxMBBAAwAQBBAAwQQABAEwQQAAAEwQQAMAEAQQAMEEAAQBMEEAAABMEEADABAEEADBBAAEATPwffXwcOwZPfVIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [9]\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:  [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:  [3]\n",
      "Label:  5\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)"
   ]
  }
 ],
 "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}