Simple_RNN_Example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Simple RNNs\n",
    "\n",
    "In this notebook, we are going to train a Simple RNNs to do a **time-series prediction**.\n",
    "\n",
    ">* Create Data\n",
    "* Define RNNs in PyTorch\n",
    "* Train the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import Library and create data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 5))\n",
    "\n",
    "#how many time step/data pts are in one batch of data\n",
    "sequence = 20\n",
    "\n",
    "# Generate Data pts\n",
    "time_steps = np.linspace(0, np.pi, sequence + 1)\n",
    "data = np.sin(time_steps)\n",
    "data.resize((sequence + 1, 1))\n",
    "\n",
    "x = data[:-1]\n",
    "y = data[1:]\n",
    "\n",
    "#plot the data points\n",
    "plt.plot(time_steps[1:], x, 'r.', label = 'input, x')  # x\n",
    "plt.plot(time_steps[1:], y, 'b.', label = 'target, y') # y\n",
    "\n",
    "plt.legend(loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define RNN using PyTorch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "class RNN(nn.Module):\n",
    "    def __init__(self, input_size, output_size, hidden_dim, n_layers):\n",
    "        super(RNN, self).__init__()\n",
    "        \n",
    "        self.hidden_dim = hidden_dim\n",
    "        \n",
    "        \n",
    "        #deine the RNN\n",
    "        self.rnn = nn.RNN(input_size, hidden_dim, n_layers, batch_first = True)\n",
    "        \n",
    "        #Fully connected layer\n",
    "        self.fc = nn.Linear(hidden_dim, output_size)\n",
    "         \n",
    "        \n",
    "    def forward(self, x, hidden):\n",
    "            batch_size = x.size(0)\n",
    "            r_out, hidden = self.rnn(x, hidden)\n",
    "            r_out = r_out.view(-1, self.hidden_dim)\n",
    "            \n",
    "            #final output\n",
    "            output = self.fc(r_out)\n",
    "            return output, hidden"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input size: torch.Size([1, 20, 1])\n",
      "Output size: torch.Size([20, 1])\n",
      "Hidden size: torch.Size([2, 1, 10])\n"
     ]
    }
   ],
   "source": [
    "#test the dimensions\n",
    "\n",
    "test_rnn = RNN( input_size=1, output_size = 1, hidden_dim = 10, n_layers =2)\n",
    "\n",
    "\n",
    "# Generate Data pts\n",
    "time_steps = np.linspace(0, np.pi, sequence + 1)\n",
    "data = np.sin(time_steps)\n",
    "data.resize((sequence, 1))\n",
    "\n",
    "test_input = torch.Tensor(data).unsqueeze(0)\n",
    "\n",
    "print ('Input size: {}'.format(test_input.size()))\n",
    "\n",
    "test_out, test_h = test_rnn(test_input, None)\n",
    "print ('Output size: {}'.format(test_out.size()))\n",
    "print ('Hidden size: {}'.format(test_h.size()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train the Model!!\n",
    "\n",
    "Let's dive in. :)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RNN(\n",
      "  (rnn): RNN(1, 32, batch_first=True)\n",
      "  (fc): Linear(in_features=32, out_features=1, bias=True)\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "# Model's Hyperparamters\n",
    "input_size = 1\n",
    "output_size = 1\n",
    "hidden_dim = 32\n",
    "n_layers = 1\n",
    "\n",
    "#instantiate a RNN\n",
    "rnn = RNN(input_size, output_size, hidden_dim, n_layers)\n",
    "print (rnn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loss and optimizer!!\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.MSELoss()\n",
    "optimizer = torch.optim.Adam(rnn.parameters(), lr = 0.01)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Define the training Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train (rnn, steps, print_every):\n",
    "    hidden = None\n",
    "    \n",
    "    for batch, step in enumerate(range(steps)):\n",
    "        time_steps = np.linspace(step * np.pi, (step+1)*np.pi, sequence + 1)\n",
    "        data = np.sin(time_steps)\n",
    "        data.resize((sequence + 1, 1))\n",
    "        \n",
    "        x = data[:-1]\n",
    "        y = data[1:]\n",
    "        \n",
    "        #Convert it into tensor\n",
    "        x_tensor = torch.Tensor(x).unsqueeze(0)\n",
    "        y_tensor = torch.Tensor(y)\n",
    "        \n",
    "        #Output\n",
    "        prediction, hidden = rnn(x_tensor, hidden)\n",
    "        \n",
    "        hidden = hidden.data\n",
    "        #Set optimizer to zero \n",
    "        optimizer.zero_grad()\n",
    "        \n",
    "        #Calculate the loss\n",
    "        loss = criterion(prediction, y_tensor)\n",
    "        \n",
    "        #perform Backprop\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "         # display loss and predictions\n",
    "        if batch%print_every == 0:        \n",
    "            print('Loss: ', loss.item())\n",
    "            plt.plot(time_steps[1:], x, 'r.') # input\n",
    "            plt.plot(time_steps[1:], prediction.data.numpy().flatten(), 'b.') # predictions\n",
    "            plt.show()\n",
    "    \n",
    "    return rnn \n",
    "        \n",
    "        \n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train it!!\n",
    "\n",
    "Finally!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.6329301595687866\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.06349427253007889\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.011008521541953087\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.002921287901699543\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.00032376620220020413\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.0003875246620737016\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.0002874383353628218\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss:  0.0002127287007169798\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": [
    "# Define some parameters\n",
    "steps= 80\n",
    "print_every = 10\n",
    "\n",
    "train_rnn = train(rnn, steps, print_every)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Time-Series Prediction\n",
    "\n",
    "Time-series prediction can be applied to many tasks. Think about weather forecasting or predicting the ebb and flow of stock market prices. You can even try to generate predictions much further in the future than just one time step!"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:root] *",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}