fundamentals_DL_OOP_from_scratch.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Red Neuronal MultiCapa, OOP model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importar Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 28, 28)\n"
     ]
    }
   ],
   "source": [
    "from get_images import get_images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# MNIST path\n",
    "mnist_path = './mnist_raw/'\n",
    "x_train_num, y_train_num, x_test_num, y_test_num = get_images(mnist_path)\n",
    "\n",
    "x_train = x_train_num[:50000].reshape(50000, -1).astype(float)\n",
    "y_train = y_train_num[:50000].reshape(50000, 1)\n",
    "\n",
    "x_val = x_train_num[50000:].reshape(10000, -1).astype(float)\n",
    "y_val = y_train_num[50000:].reshape(10000, 1)\n",
    "\n",
    "x_test = x_test_num.copy().reshape(10000, -1).astype(float)\n",
    "y_test = y_test_num.copy().reshape(10000, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(33.39512885204082, 78.6661972212754, 0.0)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.mean(), x_train.std(), x_train.min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalise(x_mean, x_std, x_data):\n",
    "    return (x_data - x_mean) / x_std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_mean = x_train.mean()\n",
    "x_std = x_train.std()\n",
    "\n",
    "x_train = normalise(x_mean, x_std, x_train)\n",
    "x_val = normalise(x_mean, x_std, x_val)\n",
    "x_test = normalise(x_mean, x_std, x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-9.646705203355238e-18, 0.9999999999999997)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.mean(), x_train.std()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Graficar muestras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_number(image):\n",
    "    plt.figure(figsize=(5,5))\n",
    "    plt.imshow(image.squeeze(), cmap=plt.get_cmap('gray'))\n",
    "    plt.axis('off')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "La imagen muestreada representa un: 8\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rnd_idx = np.random.randint(len(y_test))\n",
    "print(f'La imagen muestreada representa un: {y_test[rnd_idx, 0]}')\n",
    "plot_number(x_test_num[rnd_idx])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ecuaciones para nuestro modelo\n",
    "\n",
    "\n",
    "$$z^1 = W^1 X + b^1$$\n",
    "\n",
    "$$a^1 = ReLU(z^1) $$\n",
    "\n",
    "$$z^2 = W^2 a^1 + b^2$$\n",
    "\n",
    "$$\\hat{y} = \\frac{e^{z^{2_k}}}{\\sum_j{e^{z_j}}}$$\n",
    "\n",
    "\n",
    "$$ \\mathcal{L}(\\hat{y}^{i}, y^{i}) =  - y^{i}  \\ln(\\hat{y}^{i}) = -\\ln(\\hat{y}^i)$$\n",
    "\n",
    "\n",
    "$$ \\mathcal{J}(w, b) =  \\frac{1}{num\\_samples} \\sum_{i=1}^{num\\_samples}-\\ln(\\hat{y}^{i})$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Funciones adicionales"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Mini batches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_minibatches(mb_size, x, y, shuffle = True):\n",
    "    '''\n",
    "    x  #muestras, 784\n",
    "    y #muestras, 1\n",
    "    '''\n",
    "    assert x.shape[0] == y.shape[0], 'Error en cantidad de muestras'\n",
    "    total_data = x.shape[0]\n",
    "    if shuffle: \n",
    "        idxs = np.arange(total_data)\n",
    "        np.random.shuffle(idxs)\n",
    "        x = x[idxs]\n",
    "        y = y[idxs]  \n",
    "    return ((x[i:i+mb_size], y[i:i+mb_size]) for i in range(0, total_data, mb_size))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Nuestra clase Linear, ReLU y Sequential"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "class np_tensor(np.ndarray): pass"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Clase Linear"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "code_folding": [
     1,
     7,
     10
    ]
   },
   "outputs": [],
   "source": [
    "class Linear():\n",
    "    def __init__(self, input_size, output_size):\n",
    "        '''\n",
    "        Init parameters utilizando Kaiming He\n",
    "        '''\n",
    "        self.W = (np.random.randn(output_size, input_size) / np.sqrt(input_size/2)).view(np_tensor)\n",
    "        self.b = (np.zeros((output_size, 1))).view(np_tensor)\n",
    "    def __call__(self, X): # esta el foward de la clase lineal\n",
    "        Z = self.W @ X + self.b\n",
    "        return Z\n",
    "    def backward(self, X, Z):\n",
    "        X.grad = self.W.T @ Z.grad\n",
    "        self.W.grad = Z.grad @ X.T\n",
    "        self.b.grad = np.sum(Z.grad, axis = 1, keepdims=True)    \n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Clase ReLU"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "code_folding": [
     1,
     3
    ]
   },
   "outputs": [],
   "source": [
    "class ReLU():\n",
    "    def __call__(self, Z):\n",
    "        return np.maximum(0, Z)\n",
    "    def backward(self, Z, A):\n",
    "        Z.grad = A.grad.copy()\n",
    "        Z.grad[Z <= 0] = 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Clase Sequential"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "code_folding": [
     23
    ]
   },
   "outputs": [],
   "source": [
    "class Sequential_layers():\n",
    "    def __init__(self, layers):\n",
    "        '''\n",
    "        layers - lista que contiene objetos de tipo Linear, ReLU\n",
    "        '''\n",
    "        self.layers = layers\n",
    "        self.x = None\n",
    "        self.outputs = {}\n",
    "    def __call__(self, X):\n",
    "        self.x = X \n",
    "        self.outputs['l0'] = self.x\n",
    "        for i, layer in enumerate(self.layers, 1):\n",
    "            self.x = layer(self.x)\n",
    "            self.outputs['l'+str(i)]=self.x\n",
    "        return self.x\n",
    "    def backward(self):\n",
    "        for i in reversed(range(len(self.layers))):\n",
    "            self.layers[i].backward(self.outputs['l'+str(i)], self.outputs['l'+str(i+1)])\n",
    "    def update(self, learning_rate = 1e-3):\n",
    "        for layer in self.layers:\n",
    "            if isinstance(layer, ReLU): continue\n",
    "            layer.W = layer.W - learning_rate * layer.W.grad\n",
    "            layer.b = layer.b - learning_rate * layer.b.grad\n",
    "    def predict(self, X):\n",
    "        return np.argmax(self.__call__(X))            "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cost Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def softmaxXEntropy(x, y):\n",
    "    batch_size = x.shape[1]\n",
    "    exp_scores = np.exp(x)\n",
    "    probs = exp_scores / exp_scores.sum(axis = 0)\n",
    "    preds = probs.copy()\n",
    "    # Costo\n",
    "    y_hat = probs[y.squeeze(), np.arange(batch_size)]\n",
    "    cost = np.sum(-np.log(y_hat)) / batch_size\n",
    "    # Calcular gradientes\n",
    "    probs[y.squeeze(), np.arange(batch_size)] -= 1 #dl/dx\n",
    "    x.grad = probs.copy()\n",
    "    \n",
    "    return preds, cost"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loop de entrenamiento"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def accuracy(x, y, mb_size):\n",
    "    correct = 0\n",
    "    total = 0\n",
    "    for i, (x, y) in enumerate(create_minibatches(mb_size, x, y)):\n",
    "        pred = model(x.T.view(np_tensor))\n",
    "        correct += np.sum(np.argmax(pred, axis=0) == y.squeeze())\n",
    "        total += pred.shape[1]\n",
    "    return correct/total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "def train(model, epochs, mb_size=128, learning_rate = 1e-3):\n",
    "    for epoch in range(epochs):\n",
    "        for i, (x, y) in enumerate(create_minibatches(mb_size, x_train, y_train)):\n",
    "            scores = model(x.T.view(np_tensor))\n",
    "            _, cost = softmaxXEntropy(scores, y)\n",
    "            model.backward()\n",
    "            model.update(learning_rate)\n",
    "        print(f'costo: {cost}, accuracy: {accuracy(x_val, y_val, mb_size)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential_layers([Linear(784, 200), ReLU(), Linear(200, 200), ReLU(), Linear(200, 10)])\n",
    "mb_size = 512\n",
    "learning_rate = 1e-4\n",
    "epochs = 30\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "costo: 0.32965399257374456, accuracy: 0.9188\n",
      "costo: 0.2549719709434902, accuracy: 0.9367\n",
      "costo: 0.17610913884260815, accuracy: 0.9455\n",
      "costo: 0.22071069295077, accuracy: 0.9518\n",
      "costo: 0.16500668904844706, accuracy: 0.9542\n",
      "costo: 0.14635921694452597, accuracy: 0.9595\n",
      "costo: 0.14902383246846643, accuracy: 0.9617\n",
      "costo: 0.09150103021458665, accuracy: 0.9629\n",
      "costo: 0.13150240543171648, accuracy: 0.9647\n",
      "costo: 0.10515875498609156, accuracy: 0.9651\n",
      "costo: 0.1345903246727562, accuracy: 0.9671\n",
      "costo: 0.07315682531501502, accuracy: 0.968\n",
      "costo: 0.08936143380136623, accuracy: 0.9678\n",
      "costo: 0.10385813138229129, accuracy: 0.9686\n",
      "costo: 0.11205291679044771, accuracy: 0.9692\n",
      "costo: 0.06788661450730399, accuracy: 0.9706\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[19], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mtrain\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mepochs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmb_size\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlearning_rate\u001b[49m\u001b[43m)\u001b[49m\n",
      "Cell \u001b[0;32mIn[17], line 3\u001b[0m, in \u001b[0;36mtrain\u001b[0;34m(model, epochs, mb_size, learning_rate)\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21mtrain\u001b[39m(model, epochs, mb_size\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m128\u001b[39m, learning_rate \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1e-3\u001b[39m):\n\u001b[1;32m      2\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m epoch \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(epochs):\n\u001b[0;32m----> 3\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m i, (x, y) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(\u001b[43mcreate_minibatches\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmb_size\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my_train\u001b[49m\u001b[43m)\u001b[49m):\n\u001b[1;32m      4\u001b[0m             scores \u001b[38;5;241m=\u001b[39m model(x\u001b[38;5;241m.\u001b[39mT\u001b[38;5;241m.\u001b[39mview(np_tensor))\n\u001b[1;32m      5\u001b[0m             _, cost \u001b[38;5;241m=\u001b[39m softmaxXEntropy(scores, y)\n",
      "Cell \u001b[0;32mIn[10], line 11\u001b[0m, in \u001b[0;36mcreate_minibatches\u001b[0;34m(mb_size, x, y, shuffle)\u001b[0m\n\u001b[1;32m      9\u001b[0m     idxs \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39marange(total_data)\n\u001b[1;32m     10\u001b[0m     np\u001b[38;5;241m.\u001b[39mrandom\u001b[38;5;241m.\u001b[39mshuffle(idxs)\n\u001b[0;32m---> 11\u001b[0m     x \u001b[38;5;241m=\u001b[39m \u001b[43mx\u001b[49m\u001b[43m[\u001b[49m\u001b[43midxs\u001b[49m\u001b[43m]\u001b[49m\n\u001b[1;32m     12\u001b[0m     y \u001b[38;5;241m=\u001b[39m y[idxs]  \n\u001b[1;32m     13\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m ((x[i:i\u001b[38;5;241m+\u001b[39mmb_size], y[i:i\u001b[38;5;241m+\u001b[39mmb_size]) \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(\u001b[38;5;241m0\u001b[39m, total_data, mb_size))\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "train(model, epochs, mb_size, learning_rate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9714\n"
     ]
    }
   ],
   "source": [
    "print(accuracy(x_test, y_test, mb_size))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "el valor predicho es: 1, el valor real es:1\n"
     ]
    }
   ],
   "source": [
    "idx = np.random.randint(len(y_test))\n",
    "plot_number(x_test_num[idx])\n",
    "pred = model.predict(x_test[idx].reshape(-1, 1))\n",
    "print(f'el valor predicho es: {pred}, el valor real es:{y_test[idx][0]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.8.10"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}