sndsabin/dhcd-resnet50.ipynb

## dhcd-resnet50.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This ensures that any edits to libraries made are reloaded here automatically, and also that any charts or images displayed are shown in this notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "import fastai libraries and packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.vision import *\n",
    "from fastai.metrics import error_rate"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Looking at the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Going to use the [DHCD dataset]('https://archive.ics.uci.edu/ml/datasets/Devanagari+Handwritten+Character+Dataset') by Prashnna Kumar Gyawali et al. which features 36 devnagari alphabets and 10 devnagari digits. Model need to learn to differentiate between these 46 distinct categories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = Path('/home/sndsabin/.fastai/data/DevanagariHandwrittenCharacterDataset')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[PosixPath('/home/sndsabin/.fastai/data/DevanagariHandwrittenCharacterDataset/Test'),\n",
       " PosixPath('/home/sndsabin/.fastai/data/DevanagariHandwrittenCharacterDataset/Train')]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path.ls()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfms = get_transforms(do_flip=False)\n",
    "data = ImageDataBunch.from_folder(path, train='Train', valid='Test', ds_tfms=tfms, size=26)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.show_batch(rows=3, figsize=(5, 5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['character_10_yna', 'character_11_taamatar', 'character_12_thaa', 'character_13_daa', 'character_14_dhaa', 'character_15_adna', 'character_16_tabala', 'character_17_tha', 'character_18_da', 'character_19_dha', 'character_1_ka', 'character_20_na', 'character_21_pa', 'character_22_pha', 'character_23_ba', 'character_24_bha', 'character_25_ma', 'character_26_yaw', 'character_27_ra', 'character_28_la', 'character_29_waw', 'character_2_kha', 'character_30_motosaw', 'character_31_petchiryakha', 'character_32_patalosaw', 'character_33_ha', 'character_34_chhya', 'character_35_tra', 'character_36_gya', 'character_3_ga', 'character_4_gha', 'character_5_kna', 'character_6_cha', 'character_7_chha', 'character_8_ja', 'character_9_jha', 'digit_0', 'digit_1', 'digit_2', 'digit_3', 'digit_4', 'digit_5', 'digit_6', 'digit_7', 'digit_8', 'digit_9']\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(46, 46)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(data.classes)\n",
    "len(data.classes), data.c"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training resnet50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn = cnn_learner(data, models.resnet50, metrics=error_rate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sequential(\n",
       "  (0): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
       "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (2): ReLU(inplace)\n",
       "    (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "    (4): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "    )\n",
       "    (5): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "          (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (3): Bottleneck(\n",
       "        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "    )\n",
       "    (6): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "          (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (3): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (4): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (5): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "    )\n",
       "    (7): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "          (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace)\n",
       "      )\n",
       "    )\n",
       "  )\n",
       "  (1): Sequential(\n",
       "    (0): AdaptiveConcatPool2d(\n",
       "      (ap): AdaptiveAvgPool2d(output_size=1)\n",
       "      (mp): AdaptiveMaxPool2d(output_size=1)\n",
       "    )\n",
       "    (1): Flatten()\n",
       "    (2): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (3): Dropout(p=0.25)\n",
       "    (4): Linear(in_features=4096, out_features=512, bias=True)\n",
       "    (5): ReLU(inplace)\n",
       "    (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (7): Dropout(p=0.5)\n",
       "    (8): Linear(in_features=512, out_features=46, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "learn.model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>2.254984</td>\n",
       "      <td>1.620712</td>\n",
       "      <td>0.439058</td>\n",
       "      <td>01:47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1.279463</td>\n",
       "      <td>0.758844</td>\n",
       "      <td>0.221884</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.667283</td>\n",
       "      <td>0.308747</td>\n",
       "      <td>0.093623</td>\n",
       "      <td>01:44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.411288</td>\n",
       "      <td>0.175252</td>\n",
       "      <td>0.054638</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.278825</td>\n",
       "      <td>0.114019</td>\n",
       "      <td>0.034348</td>\n",
       "      <td>01:46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.226134</td>\n",
       "      <td>0.089676</td>\n",
       "      <td>0.026159</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.187417</td>\n",
       "      <td>0.078343</td>\n",
       "      <td>0.022464</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.165028</td>\n",
       "      <td>0.066773</td>\n",
       "      <td>0.020217</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.148426</td>\n",
       "      <td>0.058803</td>\n",
       "      <td>0.017899</td>\n",
       "      <td>01:44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.116859</td>\n",
       "      <td>0.055715</td>\n",
       "      <td>0.016377</td>\n",
       "      <td>01:47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.126697</td>\n",
       "      <td>0.054630</td>\n",
       "      <td>0.016159</td>\n",
       "      <td>01:46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>0.114919</td>\n",
       "      <td>0.053107</td>\n",
       "      <td>0.016087</td>\n",
       "      <td>01:45</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fit_one_cycle(12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn.save('stage-1-50')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unfreezing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn.unfreeze()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.223830</td>\n",
       "      <td>0.110207</td>\n",
       "      <td>0.033841</td>\n",
       "      <td>02:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.127131</td>\n",
       "      <td>0.064977</td>\n",
       "      <td>0.018841</td>\n",
       "      <td>02:14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.051933</td>\n",
       "      <td>0.036861</td>\n",
       "      <td>0.010652</td>\n",
       "      <td>02:13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.025789</td>\n",
       "      <td>0.025427</td>\n",
       "      <td>0.006739</td>\n",
       "      <td>02:13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fit_one_cycle(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save model again\n",
    "learn.save('stage-2-50')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Error Rate = 0.6739% "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "interp = ClassificationInterpretation.from_learner(learn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x792 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Top Losses\n",
    "interp.plot_top_losses(9, figsize=(15, 11))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion matrix\n",
    "interp.plot_confusion_matrix(figsize=(12, 12), dpi=60)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('character_23_ba', 'character_29_waw', 6),\n",
       " ('character_14_dhaa', 'character_18_da', 5),\n",
       " ('character_18_da', 'character_14_dhaa', 4),\n",
       " ('character_12_thaa', 'character_14_dhaa', 3),\n",
       " ('character_19_dha', 'character_4_gha', 3),\n",
       " ('character_31_petchiryakha', 'character_21_pa', 3),\n",
       " ('character_35_tra', 'character_23_ba', 3),\n",
       " ('character_10_yna', 'character_3_ga', 2),\n",
       " ('character_11_taamatar', 'character_12_thaa', 2),\n",
       " ('character_17_tha', 'character_19_dha', 2),\n",
       " ('character_17_tha', 'character_4_gha', 2),\n",
       " ('character_19_dha', 'character_17_tha', 2),\n",
       " ('character_21_pa', 'character_26_yaw', 2),\n",
       " ('character_27_ra', 'digit_2', 2),\n",
       " ('character_29_waw', 'character_16_tabala', 2),\n",
       " ('character_4_gha', 'character_26_yaw', 2)]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# most confused\n",
    "interp.most_confused(min_val=2)"
   ]
  },
  {
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}