notebooks/lin_reg.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LinReg with PyTorch, Gradient Descent, and GPU"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Init, helpers, utils ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "import torchvision\n",
    "\n",
    "DEVICE = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pprint import pprint\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from IPython.core.debugger import set_trace"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Problem"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12b1d7e48>]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.datasets import make_regression\n",
    "\n",
    "\n",
    "n_features = 1\n",
    "n_samples = 100\n",
    "\n",
    "X, y = make_regression(\n",
    "    n_samples=n_samples,\n",
    "    n_features=n_features,\n",
    "    noise=20,\n",
    "    random_state=42,\n",
    ")\n",
    "\n",
    "fix, ax = plt.subplots()\n",
    "ax.plot(X, y, \".\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Solution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = torch.from_numpy(X).float()\n",
    "y = torch.from_numpy(y.reshape((n_samples, n_features))).float()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "class LinReg(nn.Module):\n",
    "    def __init__(self, input_dim):\n",
    "        super().__init__()\n",
    "        self.beta = nn.Linear(input_dim, 1)\n",
    "        \n",
    "    def forward(self, X):\n",
    "        return self.beta(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = LinReg(n_features).to(DEVICE)  # <-- here\n",
    "loss_fn = nn.MSELoss()\n",
    "optimizer = optim.SGD(model.parameters(), lr=0.1)\n",
    "\n",
    "\n",
    "X, y = X.to(DEVICE), y.to(DEVICE)  # <-- here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No handles with labels found to put in legend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cpu\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Device\n",
    "print(DEVICE)\n",
    "\n",
    "NUM_EPOCS = 100\n",
    "\n",
    "# Train step\n",
    "# Set the model in the train model\n",
    "model.train()  # <-- here\n",
    "\n",
    "losses = []\n",
    "\n",
    "for epoch in range(NUM_EPOCS):\n",
    "    # Nulls the gradients\n",
    "    optimizer.zero_grad()\n",
    "    \n",
    "    # Do one forward pass with the data\n",
    "    y_ = model(X)\n",
    "    # calculate the loss \n",
    "    loss = loss_fn(y_, y)\n",
    "    \n",
    "    # Run a backward propagation\n",
    "    loss.backward()\n",
    "    # Optimizer updates the model parameters\n",
    "    optimizer.step()\n",
    "    losses.append(loss.item())\n",
    "\n",
    "# Visualize the output\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(losses)\n",
    "ax.set_title(\"Loss function over epochs\")\n",
    "ax.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Eval\n",
    "model.eval()  # <-- here\n",
    "with torch.no_grad():\n",
    "    y_ = model(X)    \n",
    "    \n",
    "# Visualize the output\n",
    "fig, ax = plt.subplots()\n",
    "ax.plot(X.cpu().numpy(), y_.cpu().numpy(), \".\", label=\"pred\")\n",
    "ax.plot(X.cpu().numpy(), y.cpu().numpy(), \".\", label=\"data\")\n",
    "ax.set_title(f\"MSE: {loss.item():0.1f}\")\n",
    "ax.legend();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note: I did gradient descent with all the data. I did not split the data into `train` and `valid` which should be done!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exercise:\n",
    "- Write a proper training loop for this linear regression example.\n",
    "- Split data into train and valid.\n",
    "- Use the Dataset and DataLoader abstraction.\n",
    "- Create a logistic regression module.\n",
    "- Create a Multi Layer Perceptron (MLP)."
   ]
  }
 ],
 "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}