Train_and_Validation_Losses.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from typing import Tuple, Optional\n",
    "\n",
    "import torch\n",
    "from torch import nn\n",
    "import torch.nn.functional as F\n",
    "import torchvision.transforms as transforms\n",
    "from torch.autograd import Variable\n",
    "from scipy.ndimage import gaussian_filter1d\n",
    "from tqdm.notebook import tqdm, trange\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load \"train_stats.pt\"\n",
    "Files are in the 'train_stats' directory. They have the train and validation loss values per iteration. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "losses = torch.load('./session19-shuffle/train_stats.pt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'loss': 0.008628422394394875,\n",
       " 'loss1': 0.003075739834457636,\n",
       " 'loss2': 0.005552682559937239,\n",
       " 'val_loss': 0.014393053948879242,\n",
       " 'val_loss1': 0.008403180167078972,\n",
       " 'val_loss2': 0.00598987378180027}"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "losses[100000]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Draw the loss curves "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 0.02)"
      ]
     },
     "execution_count": 6,
     "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": [
    "train_loss = []\n",
    "test_loss = []\n",
    "for i in range(1, 10001):\n",
    "    train_loss.append(losses[i]['loss'])\n",
    "\n",
    "for j in range(1, 11):\n",
    "    test_loss.append(losses[1000 * j]['val_loss'])\n",
    "    \n",
    "plt.plot(range(1, 10001), train_loss)\n",
    "plt.plot(range(1000, 10001, 1000), test_loss)\n",
    "plt.ylim([0, 0.02])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}