{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pASl9O_Gb4x1"
   },
   "source": [
    "# Introduction to Neural Networks\n",
    "\n",
    "Titouan Vayer, titouan.vayer@inria.fr \n",
    "Aurélien Garivier, aurelien.garivier@ens-lyon.fr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "2WZvU5Ocb4x5"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "sl9aeSeyb4x7"
   },
   "source": [
    "## Part 1: 1D regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "4ZQ9i0t6b4x8"
   },
   "outputs": [],
   "source": [
    "F = lambda x: 0.5*x**3-3*x**2+1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LlnxgsPJb4x9"
   },
   "source": [
    "### Computing a gradient thanks to autodiff"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XBVf_Uygb4x9"
   },
   "source": [
    "Let $F(x) = 0.5 x^{3} - 3 x^{2} +1$, and it's derivative $F'(x) = \\frac{3}{2} x^{2}- 6 x$.\n",
    "In the context of learning, the function is called by pytorch *loss*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "N1bTZQ7hb4x-"
   },
   "outputs": [],
   "source": [
    "x = torch.Tensor([2])\n",
    "x.requires_grad = True # tell pytorch it needs to compute the gradients wrt x\n",
    "loss = F(x) # computing the function value = forward pass\n",
    "loss.backward() # backward pass, to prepare the computation of the gradient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "r-ZPMonGb4x_",
    "outputId": "d3e2f506-4de4-4f29-f76e-b78dd21ed11d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-7.0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loss.item() # function value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "bL8uhoCIb4yA",
    "outputId": "4947faf0-1ad3-4e34-9863-244d80c4d0b6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([-6.])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.grad #le gradient en x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vvplZM_Qb4yB"
   },
   "source": [
    "##### Q: Check that the computed values are correct"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oGjeZamNb4yB"
   },
   "source": [
    "##### Q: Modify the code above to do the same computations at x=1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yfbWTMhtb4yE"
   },
   "source": [
    "### Simple 1D regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Gevcb_1Nb4yE"
   },
   "source": [
    "We generate random points for training. They follow a simple model \"signal + noise\", where the signal is the function $F$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "nJiYw8oqb4yF"
   },
   "outputs": [],
   "source": [
    "sigma = 50\n",
    "n = 100\n",
    "X = np.linspace(-10,10,n)\n",
    "y = F(X)+sigma*np.random.randn(n,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "dJZ_LvvJb4yF",
    "outputId": "8127e2db-1852-4779-d949-bde6f7a99e5b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(X,y,'o',lw=1, label = 'Noisy samples')\n",
    "plt.plot(X, F(X), 'k--', lw = 3, label = 'True F')\n",
    "plt.title('Generated data', fontsize = 14)\n",
    "plt.legend()\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "jE8BekwAb4yG"
   },
   "source": [
    "We use a neural network to learn, from the previously sample points, a regressor that should be as close as possible from the original function $F$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "uFSJhfAsb4yG"
   },
   "outputs": [],
   "source": [
    "class NeuralNetworkRegressor(nn.Module):\n",
    "    def __init__(self, d1 = 10):\n",
    "        super(NeuralNetworkRegressor, self).__init__()\n",
    "        self.flatten = nn.Flatten()\n",
    "        self.linear_relu_stack = nn.Sequential(\n",
    "            nn.Linear(1, d1),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(d1, d1),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(d1, d1),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(d1, 1),\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.flatten(x)\n",
    "        pred = self.linear_relu_stack(x)\n",
    "        return pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "z5PKrv42b4yH"
   },
   "source": [
    "#### Q: Comment on the previously chosen network"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "isuMhur2b4yH"
   },
   "source": [
    "In order to train our network, we need to \n",
    "  - instantiate its architecture, \n",
    "  - define an optimizer\n",
    "  - choose a learning rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "gPvZQtLIb4yH"
   },
   "outputs": [],
   "source": [
    "model = NeuralNetworkRegressor(30)\n",
    "learning_rate = 5e-2\n",
    "epochs = 100\n",
    "\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)\n",
    "Xt = torch.from_numpy(X).type(torch.float).reshape(-1,1) #pytorch manipulates only tensors, not numpy array\n",
    "yt = torch.from_numpy(y).type(torch.float).reshape(-1,1)\n",
    "loss_fn = nn.MSELoss() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "7vfgBQTzb4yI"
   },
   "outputs": [],
   "source": [
    "pytorch_total_params = sum(p.numel() for p in model.parameters())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "xhZGpjgPb4yI",
    "outputId": "c280d159-4b3e-4ae0-d473-c12d5652a1c1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1951\n"
     ]
    }
   ],
   "source": [
    "print(pytorch_total_params) # total number of parameters of the model: comment!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "siV3tSkcb4yI"
   },
   "source": [
    "#### Q: comment on the choice of the loss function"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "V2BAHdQJb4yI"
   },
   "source": [
    "We can train the network by using the following loop:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "0EHQKTbAb4yJ"
   },
   "outputs": [],
   "source": [
    "losses = []\n",
    "for i in range(epochs):\n",
    "    # Compute prediction and loss\n",
    "    pred = model(Xt)\n",
    "    loss = loss_fn(pred, yt)\n",
    "\n",
    "    # Backpropagation\n",
    "    optimizer.zero_grad()\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "    losses.append(loss.item())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YUHNR8z1b4yJ"
   },
   "source": [
    "We can now display the evolution of the training error as a function of the iterations: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "M5ZFnPFAb4yJ",
    "outputId": "932e57e5-1a8c-44bb-e761-4409526f9377"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Iterations')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.semilogy(losses, lw = 2)\n",
    "plt.title('Training loss', fontsize = 14)\n",
    "plt.xlabel('Iterations', fontsize = 14)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KrPvkrsWb4yK"
   },
   "source": [
    "#### Q: Did the algorithm converge?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "D_ScZSJYb4yK"
   },
   "source": [
    "Let us now show the model prediction, and compare with the original function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "b0uSBzgTb4yK",
    "outputId": "48e152d1-286a-4d9c-ea4a-ac6f62dcc238"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(X,y,'o',lw=1)\n",
    "yhat = model(torch.from_numpy(X).type(torch.float).reshape(-1,1)).detach().numpy()\n",
    "plt.plot(X,yhat,'r-',lw=2, label = 'prediction')\n",
    "plt.plot(X, F(X), 'k--', lw = 3, label = 'True F')\n",
    "plt.title('Number of params = {}'.format(pytorch_total_params))\n",
    "plt.legend()\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bN_lkuWub4yK"
   },
   "source": [
    "#### Q: Comment on the previous experiment. How to validate properly the fitted model?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0VL4QR0gb4yL"
   },
   "source": [
    "#### Q: What happens when the number of parameters explodes ? (for example, take $d_1 = 200$)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "2rALieR2b4yL"
   },
   "outputs": [],
   "source": [
    "model2 = NeuralNetworkRegressor(200)\n",
    "learning_rate = 5e-2\n",
    "epochs = 200\n",
    "\n",
    "optimizer = torch.optim.Adam(model2.parameters(), lr=learning_rate)\n",
    "Xt = torch.from_numpy(X).type(torch.float).reshape(-1,1) \n",
    "yt = torch.from_numpy(y).type(torch.float).reshape(-1,1)\n",
    "loss_fn = nn.MSELoss() \n",
    "pytorch_total_params2 = sum(p.numel() for p in model2.parameters())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "0KoMOqznb4yM"
   },
   "outputs": [],
   "source": [
    "losses = []\n",
    "for i in range(epochs):\n",
    "    # Compute prediction and loss\n",
    "    pred = model2(Xt)\n",
    "    loss = loss_fn(pred, yt)\n",
    "\n",
    "    # Backpropagation\n",
    "    optimizer.zero_grad()\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "    losses.append(loss.item())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "dqkEbOPcb4yM",
    "outputId": "2c64a594-1b8d-4033-d31c-1fce7d99118e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Iterations')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.semilogy(losses, lw = 2)\n",
    "plt.title('Training loss', fontsize = 14)\n",
    "plt.xlabel('Iterations', fontsize = 14)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "guQ3XGtvb4yM",
    "outputId": "5e58a7fa-6a5b-4c2f-a38d-e04ab0721ed3"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(X,y,'o',lw=1)\n",
    "yhat = model2(torch.from_numpy(X).type(torch.float).reshape(-1,1)).detach().numpy()\n",
    "plt.plot(X,yhat,'r-',lw=2, label = 'prediction')\n",
    "plt.plot(X, F(X), 'k--', lw = 3, label = 'True F')\n",
    "plt.title('Number of params = {}'.format(pytorch_total_params2))\n",
    "plt.legend()\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}
