{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "dad342cc-22c6-4cfe-b217-cf836ea4c1fd",
   "metadata": {},
   "source": [
    "# Sample Integrated"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c27799cf-0328-4e76-9a46-6236907e1bf8",
   "metadata": {},
   "source": [
    "This notebook will do a simple pipeline based on fastai with transfer learning concepts using resnet architecture.\n",
    "Typical ideas I want to implement here \n",
    "1. Get a working pipeline with the current dataset ( using Fastai)\n",
    "2. Do a submission on Kaggle\n",
    "3. Do some iterations using simple data augmentations\n",
    "\n",
    "We will split the train dataset into 80-20 split. The idea is to have an actual measurement of model performance looking at the training data. However, trick for kaggle here is to train the model on entire training dataset before submitting results and reviewing score on the test. \n",
    "\n",
    "```{note}\n",
    "- Big Images are not easy to fit on GPU.\n",
    "- Resizing big image to smaller sizes tends to loose certain features\n",
    "- Next ideas \n",
    "    - Apply Digit Cleaner concept\n",
    "    - Create Clean Dataset / Visualize and review accuracy of digit cleaner\n",
    "    - Figure out way to do a bounding box\n",
    "    - After that we can apply a few ideas\n",
    "        - Apply resnet after digit cleaner ( current data had 0.93 error_rate after 1 epoch)\n",
    "        - Do bounding box and break down dataset into individual digits. Then do clustering to 10 categories. Then use labels to calculate sum_digit\n",
    "        - Split into individual digit images -> Resize -> Merge (permutations) into single image then train model on new dataset\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c371c7e-f822-43eb-ba21-43f8c4afc7c7",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "38ef8069-d896-46fe-b0aa-ad81e9765320",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "382128cb-e5eb-4074-b93e-6dc31a9b3818",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tue Mar 15 09:43:54 2022       \n",
      "+-----------------------------------------------------------------------------+\n",
      "| NVIDIA-SMI 470.103.01   Driver Version: 470.103.01   CUDA Version: 11.4     |\n",
      "|-------------------------------+----------------------+----------------------+\n",
      "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
      "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
      "|                               |                      |               MIG M. |\n",
      "|===============================+======================+======================|\n",
      "|   0  Tesla K80           Off  | 00000000:00:1E.0 Off |                    0 |\n",
      "| N/A   67C    P0    61W / 149W |   1316MiB / 11441MiB |      0%      Default |\n",
      "|                               |                      |                  N/A |\n",
      "+-------------------------------+----------------------+----------------------+\n",
      "                                                                               \n",
      "+-----------------------------------------------------------------------------+\n",
      "| Processes:                                                                  |\n",
      "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
      "|        ID   ID                                                   Usage      |\n",
      "|=============================================================================|\n",
      "|    0   N/A  N/A    173195      C   ...da/envs/aiking/bin/python     1313MiB |\n",
      "+-----------------------------------------------------------------------------+\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "torch.cuda.empty_cache()\n",
    "!nvidia-smi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1ddf8de9-a744-4967-9bbf-333988f389e0",
   "metadata": {},
   "outputs": [],
   "source": [
    "import fastai\n",
    "from fastai.vision.all import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2e5c0589-e70a-41ea-8295-86e14a7a9488",
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.vision.all import *\n",
    "from fastai.vision.widgets import *\n",
    "from aiking.data.external import * #We need to import this after fastai modules\n",
    "import warnings\n",
    "from matplotlib import cm\n",
    "import shutil\n",
    "import pathlib\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "243de8eb-c082-423b-8e65-4f710dc28ee4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized\n",
       "1                                  /Landmark2/pdo/aiking/data/ultra-mnist/test_black\n",
       "2     /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated\n",
       "3                            /Landmark2/pdo/aiking/data/ultra-mnist/train_sample.csv\n",
       "4                                  /Landmark2/pdo/aiking/data/ultra-mnist/sample.csv\n",
       "5                                   /Landmark2/pdo/aiking/data/ultra-mnist/train.csv\n",
       "6                                      /Landmark2/pdo/aiking/data/ultra-mnist/models\n",
       "7                                       /Landmark2/pdo/aiking/data/ultra-mnist/train\n",
       "8                             /Landmark2/pdo/aiking/data/ultra-mnist/train_train.csv\n",
       "9                                /Landmark2/pdo/aiking/data/ultra-mnist/sample_black\n",
       "10                                       /Landmark2/pdo/aiking/data/ultra-mnist/test\n",
       "11                               /Landmark2/pdo/aiking/data/ultra-mnist/train_black2\n",
       "12                       /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped\n",
       "13                            /Landmark2/pdo/aiking/data/ultra-mnist/valid_train.csv\n",
       "14                            /Landmark2/pdo/aiking/data/ultra-mnist/ultra-mnist.zip\n",
       "15                      /Landmark2/pdo/aiking/data/ultra-mnist/sample_submission.csv\n",
       "16                                /Landmark2/pdo/aiking/data/ultra-mnist/train_black\n",
       "17                                /Landmark2/pdo/aiking/data/ultra-mnist/test_black2\n",
       "18                                     /Landmark2/pdo/aiking/data/ultra-mnist/sample\n",
       "19                            /Landmark2/pdo/aiking/data/ultra-mnist/sample_intg.csv\n",
       "20                           /Landmark2/pdo/aiking/data/ultra-mnist/valid_sample.csv\n",
       "dtype: object"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = untar_data(\"kaggle_competitions::ultra-mnist\"); path\n",
    "\n",
    "pd.Series((path).ls())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7aebf2c5-e8bf-443e-8b2f-b1dc24b5bb55",
   "metadata": {},
   "outputs": [],
   "source": [
    "# !mkdir {path}/'sample'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a249be32-c18b-4b55-b8df-dd5e51043e9e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated')"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path/'sample_black_cropped_resized_integrated'\n",
    "# path/'train'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5126bcc1-a0cf-49d0-8a23-75219e2b5292",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(#560) [Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/uznkagrqln'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/gkuhpigoxr'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvkpbnawop'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/jfsumbgtsg'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/efstasnumi'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/dhxwgmgutm'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/xbgfseogty'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/ytlwuwflrt'),Path('/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/nphunajnco')...]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(path/\"sample_black_cropped_resized_integrated\").ls()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "938e33f3-244f-451b-bb6f-0e273c7b9d3d",
   "metadata": {},
   "source": [
    "## Form new train and valid df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "75666849-ab30-4211-a2c7-e33874c8b288",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>fname</th>\n",
       "      <th>id</th>\n",
       "      <th>dstype</th>\n",
       "      <th>digit_sum</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_doocaoebo.jpeg</td>\n",
       "      <td>mptgfndbad_doocaoebo.jpeg</td>\n",
       "      <td>mptgfndbad</td>\n",
       "      <td>train</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_ocboeoaod.jpeg</td>\n",
       "      <td>mptgfndbad_ocboeoaod.jpeg</td>\n",
       "      <td>mptgfndbad</td>\n",
       "      <td>train</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_cdaoooebo.jpeg</td>\n",
       "      <td>mptgfndbad_cdaoooebo.jpeg</td>\n",
       "      <td>mptgfndbad</td>\n",
       "      <td>train</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_bacoooeod.jpeg</td>\n",
       "      <td>mptgfndbad_bacoooeod.jpeg</td>\n",
       "      <td>mptgfndbad</td>\n",
       "      <td>train</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_oocabodoe.jpeg</td>\n",
       "      <td>mptgfndbad_oocabodoe.jpeg</td>\n",
       "      <td>mptgfndbad</td>\n",
       "      <td>train</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5029</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_dcboaoooo.jpeg</td>\n",
       "      <td>tvqvvpnnpf_dcboaoooo.jpeg</td>\n",
       "      <td>tvqvvpnnpf</td>\n",
       "      <td>train</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5030</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_adcoooobo.jpeg</td>\n",
       "      <td>tvqvvpnnpf_adcoooobo.jpeg</td>\n",
       "      <td>tvqvvpnnpf</td>\n",
       "      <td>train</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5031</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_ooocodboa.jpeg</td>\n",
       "      <td>tvqvvpnnpf_ooocodboa.jpeg</td>\n",
       "      <td>tvqvvpnnpf</td>\n",
       "      <td>train</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5032</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_oocaoodob.jpeg</td>\n",
       "      <td>tvqvvpnnpf_oocaoodob.jpeg</td>\n",
       "      <td>tvqvvpnnpf</td>\n",
       "      <td>train</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5033</th>\n",
       "      <td>/Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_oaodoocbo.jpeg</td>\n",
       "      <td>tvqvvpnnpf_oaodoocbo.jpeg</td>\n",
       "      <td>tvqvvpnnpf</td>\n",
       "      <td>train</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5034 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                     path  \\\n",
       "0     /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_doocaoebo.jpeg   \n",
       "1     /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_ocboeoaod.jpeg   \n",
       "2     /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_cdaoooebo.jpeg   \n",
       "3     /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_bacoooeod.jpeg   \n",
       "4     /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/mptgfndbad/mptgfndbad_oocabodoe.jpeg   \n",
       "...                                                                                                                   ...   \n",
       "5029  /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_dcboaoooo.jpeg   \n",
       "5030  /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_adcoooobo.jpeg   \n",
       "5031  /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_ooocodboa.jpeg   \n",
       "5032  /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_oocaoodob.jpeg   \n",
       "5033  /Landmark2/pdo/aiking/data/ultra-mnist/sample_black_cropped_resized_integrated/tvqvvpnnpf/tvqvvpnnpf_oaodoocbo.jpeg   \n",
       "\n",
       "                          fname          id dstype  digit_sum  \n",
       "0     mptgfndbad_doocaoebo.jpeg  mptgfndbad  train          9  \n",
       "1     mptgfndbad_ocboeoaod.jpeg  mptgfndbad  train          9  \n",
       "2     mptgfndbad_cdaoooebo.jpeg  mptgfndbad  train          9  \n",
       "3     mptgfndbad_bacoooeod.jpeg  mptgfndbad  train          9  \n",
       "4     mptgfndbad_oocabodoe.jpeg  mptgfndbad  train          9  \n",
       "...                         ...         ...    ...        ...  \n",
       "5029  tvqvvpnnpf_dcboaoooo.jpeg  tvqvvpnnpf  train         21  \n",
       "5030  tvqvvpnnpf_adcoooobo.jpeg  tvqvvpnnpf  train         21  \n",
       "5031  tvqvvpnnpf_ooocodboa.jpeg  tvqvvpnnpf  train         21  \n",
       "5032  tvqvvpnnpf_oocaoodob.jpeg  tvqvvpnnpf  train         21  \n",
       "5033  tvqvvpnnpf_oaodoocbo.jpeg  tvqvvpnnpf  train         21  \n",
       "\n",
       "[5034 rows x 5 columns]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# df_train = pd.read_csv(path/'train_sample.csv')\n",
    "# df_train['id'] = df_train['id'] + \".jpeg\"\n",
    "# df_train\n",
    "\n",
    "df_sample_intg = pd.read_csv(path/\"sample_intg.csv\"); df_sample_intg\n",
    "df_sample_train = pd.read_csv(path/\"train_sample.csv\"); df_sample_train\n",
    "df_sample_valid = pd.read_csv(path/\"valid_sample.csv\"); df_sample_valid\n",
    "df_sample = pd.concat([df_sample_train, df_sample_valid])\n",
    "# df_sample_intg['digit_sum'] = None\n",
    "df_sample_intg['dstype'] = None\n",
    "# df_sample_intg.shape, df_sample_intg.join(df_sample_train.set_index('id'), on='id').shape\n",
    "train_filter = df_sample_intg['id'].isin(df_sample_train.id)\n",
    "valid_filter = df_sample_intg['id'].isin(df_sample_valid.id)\n",
    "df_sample_intg.loc[train_filter, 'dstype'] = 'train'\n",
    "df_sample_intg.loc[valid_filter, 'dstype'] = 'valid'\n",
    "df_sample_intg['digit_sum'] = df_sample_intg.join(df_sample.set_index('id'), on='id')['digit_sum']\n",
    "df_sample_intg\n",
    "# df_sample_intg.loc[valid_filter, 'digit_sum'] = df_sample_intg[valid_filter].join(df_sample_valid.set_index('id'), on='id')['digit_sum']\n",
    "# df_sample_intg.dtypes, df_sample_train.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c735b18c-365c-42fb-b1bc-ed14517df5bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m\n",
       "\u001b[0mImageDataLoaders\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_df\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mdf\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mpath\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'.'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mvalid_pct\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.2\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mseed\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mfn_col\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mfolder\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0msuff\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m''\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mlabel_col\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mlabel_delim\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0my_block\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mvalid_col\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mitem_tfms\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mbatch_tfms\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mbs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m64\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mval_bs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mshuffle\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mdevice\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mDocstring:\u001b[0m Create from `df` using `fn_col` and `label_col`\n",
       "\u001b[0;31mFile:\u001b[0m      /opt/anaconda/envs/aiking/lib/python3.9/site-packages/fastai/vision/data.py\n",
       "\u001b[0;31mType:\u001b[0m      method\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "ImageDataLoaders.from_df?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9ec149a0-41d3-42d7-a6f5-ac3d6d8f3abc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>fname</th>\n",
       "      <th>id</th>\n",
       "      <th>dstype</th>\n",
       "      <th>digit_sum</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_aoodooecb.jpeg</td>\n",
       "      <td>jfsumbgtsg_aoodooecb.jpeg</td>\n",
       "      <td>jfsumbgtsg</td>\n",
       "      <td>valid</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_oaooobecd.jpeg</td>\n",
       "      <td>jfsumbgtsg_oaooobecd.jpeg</td>\n",
       "      <td>jfsumbgtsg</td>\n",
       "      <td>valid</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_oobcaeood.jpeg</td>\n",
       "      <td>jfsumbgtsg_oobcaeood.jpeg</td>\n",
       "      <td>jfsumbgtsg</td>\n",
       "      <td>valid</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_doeocboao.jpeg</td>\n",
       "      <td>jfsumbgtsg_doeocboao.jpeg</td>\n",
       "      <td>jfsumbgtsg</td>\n",
       "      <td>valid</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_oaoobcode.jpeg</td>\n",
       "      <td>jfsumbgtsg_oaoobcode.jpeg</td>\n",
       "      <td>jfsumbgtsg</td>\n",
       "      <td>valid</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4957</th>\n",
       "      <td>sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_obafcooed.jpeg</td>\n",
       "      <td>mdggejfmfw_obafcooed.jpeg</td>\n",
       "      <td>mdggejfmfw</td>\n",
       "      <td>valid</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4958</th>\n",
       "      <td>sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_abdoeofco.jpeg</td>\n",
       "      <td>mdggejfmfw_abdoeofco.jpeg</td>\n",
       "      <td>mdggejfmfw</td>\n",
       "      <td>valid</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4959</th>\n",
       "      <td>sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_bcefaoodo.jpeg</td>\n",
       "      <td>mdggejfmfw_bcefaoodo.jpeg</td>\n",
       "      <td>mdggejfmfw</td>\n",
       "      <td>valid</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4960</th>\n",
       "      <td>sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_eafoobdco.jpeg</td>\n",
       "      <td>mdggejfmfw_eafoobdco.jpeg</td>\n",
       "      <td>mdggejfmfw</td>\n",
       "      <td>valid</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4961</th>\n",
       "      <td>sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_bocaoodef.jpeg</td>\n",
       "      <td>mdggejfmfw_bocaoodef.jpeg</td>\n",
       "      <td>mdggejfmfw</td>\n",
       "      <td>valid</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1007 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                              path  \\\n",
       "36    sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_aoodooecb.jpeg   \n",
       "37    sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_oaooobecd.jpeg   \n",
       "38    sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_oobcaeood.jpeg   \n",
       "39    sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_doeocboao.jpeg   \n",
       "40    sample_black_cropped_resized_integrated/jfsumbgtsg/jfsumbgtsg_oaoobcode.jpeg   \n",
       "...                                                                            ...   \n",
       "4957  sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_obafcooed.jpeg   \n",
       "4958  sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_abdoeofco.jpeg   \n",
       "4959  sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_bcefaoodo.jpeg   \n",
       "4960  sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_eafoobdco.jpeg   \n",
       "4961  sample_black_cropped_resized_integrated/mdggejfmfw/mdggejfmfw_bocaoodef.jpeg   \n",
       "\n",
       "                          fname          id dstype  digit_sum  \n",
       "36    jfsumbgtsg_aoodooecb.jpeg  jfsumbgtsg  valid          9  \n",
       "37    jfsumbgtsg_oaooobecd.jpeg  jfsumbgtsg  valid          9  \n",
       "38    jfsumbgtsg_oobcaeood.jpeg  jfsumbgtsg  valid          9  \n",
       "39    jfsumbgtsg_doeocboao.jpeg  jfsumbgtsg  valid          9  \n",
       "40    jfsumbgtsg_oaoobcode.jpeg  jfsumbgtsg  valid          9  \n",
       "...                         ...         ...    ...        ...  \n",
       "4957  mdggejfmfw_obafcooed.jpeg  mdggejfmfw  valid         16  \n",
       "4958  mdggejfmfw_abdoeofco.jpeg  mdggejfmfw  valid         16  \n",
       "4959  mdggejfmfw_bcefaoodo.jpeg  mdggejfmfw  valid         16  \n",
       "4960  mdggejfmfw_eafoobdco.jpeg  mdggejfmfw  valid         16  \n",
       "4961  mdggejfmfw_bocaoodef.jpeg  mdggejfmfw  valid         16  \n",
       "\n",
       "[1007 rows x 5 columns]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sample_intg['path']= df_sample_intg['path'].apply(lambda p : pathlib.Path(p).relative_to(path))\n",
    "df_train = df_sample_intg[df_sample_intg['dstype'] =='train']\n",
    "df_valid = df_sample_intg[df_sample_intg['dstype'] =='valid']\n",
    "df_valid\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2a72cf2f-31b9-4c1b-b2b1-be110ba7e915",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<fastai.data.core.DataLoaders at 0x152ef20e5760>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dls = ImageDataLoaders.from_df(df_train,path, folder=None, valid_pct=0.2, fn_col=0, label_col=4, bs=16); dls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4a4a8073-4f12-4358-8f26-919ddd96212c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "aded5a22-eb3a-417d-bee9-ed68b248a573",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dls.show_batch()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cd2f0714-b0c4-469e-a9a4-5bb7a41e3755",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tue Mar 15 09:44:09 2022       \n",
      "+-----------------------------------------------------------------------------+\n",
      "| NVIDIA-SMI 470.103.01   Driver Version: 470.103.01   CUDA Version: 11.4     |\n",
      "|-------------------------------+----------------------+----------------------+\n",
      "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
      "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
      "|                               |                      |               MIG M. |\n",
      "|===============================+======================+======================|\n",
      "|   0  Tesla K80           Off  | 00000000:00:1E.0 Off |                    0 |\n",
      "| N/A   68C    P0    62W / 149W |   1316MiB / 11441MiB |      0%      Default |\n",
      "|                               |                      |                  N/A |\n",
      "+-------------------------------+----------------------+----------------------+\n",
      "                                                                               \n",
      "+-----------------------------------------------------------------------------+\n",
      "| Processes:                                                                  |\n",
      "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
      "|        ID   ID                                                   Usage      |\n",
      "|=============================================================================|\n",
      "|    0   N/A  N/A    173195      C   ...da/envs/aiking/bin/python     1313MiB |\n",
      "+-----------------------------------------------------------------------------+\n"
     ]
    }
   ],
   "source": [
    "!nvidia-smi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b84d620f-68d6-4ab1-acc9-0987e205bc65",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<fastai.learner.Learner at 0x152edfc161c0>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "learn = cnn_learner(dls, resnet34, metrics=[error_rate, accuracy]); learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c80263b2-01d7-4972-80e4-410a66075b74",
   "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>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>4.799863</td>\n",
       "      <td>3.852269</td>\n",
       "      <td>0.934161</td>\n",
       "      <td>0.065839</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>4.440562</td>\n",
       "      <td>3.630288</td>\n",
       "      <td>0.913043</td>\n",
       "      <td>0.086957</td>\n",
       "      <td>00:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>4.137702</td>\n",
       "      <td>3.485804</td>\n",
       "      <td>0.910559</td>\n",
       "      <td>0.089441</td>\n",
       "      <td>00:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>3.930142</td>\n",
       "      <td>3.326438</td>\n",
       "      <td>0.881988</td>\n",
       "      <td>0.118012</td>\n",
       "      <td>00:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>3.772889</td>\n",
       "      <td>3.232335</td>\n",
       "      <td>0.875776</td>\n",
       "      <td>0.124224</td>\n",
       "      <td>00:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>3.537516</td>\n",
       "      <td>3.166939</td>\n",
       "      <td>0.872050</td>\n",
       "      <td>0.127950</td>\n",
       "      <td>00:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>3.261934</td>\n",
       "      <td>3.110428</td>\n",
       "      <td>0.859627</td>\n",
       "      <td>0.140373</td>\n",
       "      <td>00:18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>3.226398</td>\n",
       "      <td>3.069852</td>\n",
       "      <td>0.857143</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>3.125710</td>\n",
       "      <td>3.024059</td>\n",
       "      <td>0.842236</td>\n",
       "      <td>0.157764</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>2.930925</td>\n",
       "      <td>3.010297</td>\n",
       "      <td>0.827329</td>\n",
       "      <td>0.172671</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>2.793818</td>\n",
       "      <td>2.934250</td>\n",
       "      <td>0.831056</td>\n",
       "      <td>0.168944</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>2.674159</td>\n",
       "      <td>2.938103</td>\n",
       "      <td>0.821118</td>\n",
       "      <td>0.178882</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>2.607552</td>\n",
       "      <td>2.868927</td>\n",
       "      <td>0.804969</td>\n",
       "      <td>0.195031</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>2.445592</td>\n",
       "      <td>2.868701</td>\n",
       "      <td>0.811180</td>\n",
       "      <td>0.188820</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>2.324407</td>\n",
       "      <td>2.838203</td>\n",
       "      <td>0.796273</td>\n",
       "      <td>0.203727</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>2.280806</td>\n",
       "      <td>2.842952</td>\n",
       "      <td>0.806211</td>\n",
       "      <td>0.193789</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>2.259488</td>\n",
       "      <td>2.829938</td>\n",
       "      <td>0.780124</td>\n",
       "      <td>0.219876</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>2.162574</td>\n",
       "      <td>2.797639</td>\n",
       "      <td>0.763975</td>\n",
       "      <td>0.236025</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>2.081153</td>\n",
       "      <td>2.750004</td>\n",
       "      <td>0.751553</td>\n",
       "      <td>0.248447</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>1.971656</td>\n",
       "      <td>2.738195</td>\n",
       "      <td>0.757764</td>\n",
       "      <td>0.242236</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>1.952935</td>\n",
       "      <td>2.759722</td>\n",
       "      <td>0.786335</td>\n",
       "      <td>0.213665</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>1.873172</td>\n",
       "      <td>2.714657</td>\n",
       "      <td>0.762733</td>\n",
       "      <td>0.237267</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>1.861983</td>\n",
       "      <td>2.709193</td>\n",
       "      <td>0.734161</td>\n",
       "      <td>0.265839</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>1.857842</td>\n",
       "      <td>2.599679</td>\n",
       "      <td>0.727950</td>\n",
       "      <td>0.272050</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>1.770387</td>\n",
       "      <td>2.656592</td>\n",
       "      <td>0.734161</td>\n",
       "      <td>0.265839</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>1.717701</td>\n",
       "      <td>2.600894</td>\n",
       "      <td>0.732919</td>\n",
       "      <td>0.267081</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>1.651577</td>\n",
       "      <td>2.582834</td>\n",
       "      <td>0.696894</td>\n",
       "      <td>0.303106</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>1.646941</td>\n",
       "      <td>2.565307</td>\n",
       "      <td>0.719255</td>\n",
       "      <td>0.280745</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>1.605862</td>\n",
       "      <td>2.606378</td>\n",
       "      <td>0.709317</td>\n",
       "      <td>0.290683</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>29</td>\n",
       "      <td>1.578137</td>\n",
       "      <td>2.604069</td>\n",
       "      <td>0.690683</td>\n",
       "      <td>0.309317</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>1.487164</td>\n",
       "      <td>2.589584</td>\n",
       "      <td>0.688199</td>\n",
       "      <td>0.311801</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>1.499904</td>\n",
       "      <td>2.622078</td>\n",
       "      <td>0.685714</td>\n",
       "      <td>0.314286</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>32</td>\n",
       "      <td>1.416947</td>\n",
       "      <td>2.607370</td>\n",
       "      <td>0.695652</td>\n",
       "      <td>0.304348</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>33</td>\n",
       "      <td>1.483090</td>\n",
       "      <td>2.651856</td>\n",
       "      <td>0.689441</td>\n",
       "      <td>0.310559</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>34</td>\n",
       "      <td>1.375965</td>\n",
       "      <td>2.557366</td>\n",
       "      <td>0.667081</td>\n",
       "      <td>0.332919</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>35</td>\n",
       "      <td>1.371571</td>\n",
       "      <td>2.497138</td>\n",
       "      <td>0.633540</td>\n",
       "      <td>0.366460</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>36</td>\n",
       "      <td>1.211827</td>\n",
       "      <td>2.438924</td>\n",
       "      <td>0.645963</td>\n",
       "      <td>0.354037</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37</td>\n",
       "      <td>1.239312</td>\n",
       "      <td>2.473795</td>\n",
       "      <td>0.618634</td>\n",
       "      <td>0.381366</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>38</td>\n",
       "      <td>1.246859</td>\n",
       "      <td>2.457524</td>\n",
       "      <td>0.619876</td>\n",
       "      <td>0.380124</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39</td>\n",
       "      <td>1.235076</td>\n",
       "      <td>2.392699</td>\n",
       "      <td>0.612422</td>\n",
       "      <td>0.387578</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>1.121472</td>\n",
       "      <td>2.471771</td>\n",
       "      <td>0.634783</td>\n",
       "      <td>0.365217</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41</td>\n",
       "      <td>1.137847</td>\n",
       "      <td>2.335734</td>\n",
       "      <td>0.627329</td>\n",
       "      <td>0.372671</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>42</td>\n",
       "      <td>1.139235</td>\n",
       "      <td>2.393010</td>\n",
       "      <td>0.619876</td>\n",
       "      <td>0.380124</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>43</td>\n",
       "      <td>1.119598</td>\n",
       "      <td>2.397207</td>\n",
       "      <td>0.611180</td>\n",
       "      <td>0.388820</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>44</td>\n",
       "      <td>1.060512</td>\n",
       "      <td>2.402379</td>\n",
       "      <td>0.614907</td>\n",
       "      <td>0.385093</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45</td>\n",
       "      <td>1.056118</td>\n",
       "      <td>2.368381</td>\n",
       "      <td>0.598758</td>\n",
       "      <td>0.401242</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>46</td>\n",
       "      <td>1.030649</td>\n",
       "      <td>2.285016</td>\n",
       "      <td>0.583851</td>\n",
       "      <td>0.416149</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>47</td>\n",
       "      <td>0.960971</td>\n",
       "      <td>2.176330</td>\n",
       "      <td>0.563975</td>\n",
       "      <td>0.436025</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>48</td>\n",
       "      <td>1.022968</td>\n",
       "      <td>2.215484</td>\n",
       "      <td>0.582609</td>\n",
       "      <td>0.417391</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>49</td>\n",
       "      <td>0.937664</td>\n",
       "      <td>2.248081</td>\n",
       "      <td>0.550311</td>\n",
       "      <td>0.449689</td>\n",
       "      <td>00:17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.867752</td>\n",
       "      <td>2.070304</td>\n",
       "      <td>0.532919</td>\n",
       "      <td>0.467081</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.657931</td>\n",
       "      <td>1.962824</td>\n",
       "      <td>0.499379</td>\n",
       "      <td>0.500621</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.491939</td>\n",
       "      <td>1.833532</td>\n",
       "      <td>0.460870</td>\n",
       "      <td>0.539130</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.429447</td>\n",
       "      <td>1.721688</td>\n",
       "      <td>0.448447</td>\n",
       "      <td>0.551553</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.359826</td>\n",
       "      <td>1.722428</td>\n",
       "      <td>0.438509</td>\n",
       "      <td>0.561491</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.301921</td>\n",
       "      <td>1.702332</td>\n",
       "      <td>0.439752</td>\n",
       "      <td>0.560248</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.297945</td>\n",
       "      <td>1.634223</td>\n",
       "      <td>0.429814</td>\n",
       "      <td>0.570186</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.272402</td>\n",
       "      <td>1.613680</td>\n",
       "      <td>0.423602</td>\n",
       "      <td>0.576398</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.217226</td>\n",
       "      <td>1.498233</td>\n",
       "      <td>0.408696</td>\n",
       "      <td>0.591304</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.192630</td>\n",
       "      <td>1.502558</td>\n",
       "      <td>0.407453</td>\n",
       "      <td>0.592547</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.168445</td>\n",
       "      <td>1.563167</td>\n",
       "      <td>0.424845</td>\n",
       "      <td>0.575155</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>0.213405</td>\n",
       "      <td>1.507696</td>\n",
       "      <td>0.375155</td>\n",
       "      <td>0.624845</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0.232360</td>\n",
       "      <td>1.438052</td>\n",
       "      <td>0.355280</td>\n",
       "      <td>0.644720</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0.195828</td>\n",
       "      <td>1.398571</td>\n",
       "      <td>0.345342</td>\n",
       "      <td>0.654658</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0.187390</td>\n",
       "      <td>1.284458</td>\n",
       "      <td>0.339130</td>\n",
       "      <td>0.660870</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>0.185759</td>\n",
       "      <td>1.265104</td>\n",
       "      <td>0.341615</td>\n",
       "      <td>0.658385</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0.201981</td>\n",
       "      <td>1.307698</td>\n",
       "      <td>0.325466</td>\n",
       "      <td>0.674534</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>0.164799</td>\n",
       "      <td>1.397388</td>\n",
       "      <td>0.350311</td>\n",
       "      <td>0.649689</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0.233455</td>\n",
       "      <td>1.324180</td>\n",
       "      <td>0.322981</td>\n",
       "      <td>0.677019</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>0.197242</td>\n",
       "      <td>1.124726</td>\n",
       "      <td>0.299379</td>\n",
       "      <td>0.700621</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>0.194921</td>\n",
       "      <td>1.244918</td>\n",
       "      <td>0.318012</td>\n",
       "      <td>0.681988</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>0.232981</td>\n",
       "      <td>1.216448</td>\n",
       "      <td>0.298137</td>\n",
       "      <td>0.701863</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>0.208103</td>\n",
       "      <td>1.070318</td>\n",
       "      <td>0.267081</td>\n",
       "      <td>0.732919</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>0.190553</td>\n",
       "      <td>1.139177</td>\n",
       "      <td>0.304348</td>\n",
       "      <td>0.695652</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>0.195688</td>\n",
       "      <td>1.171082</td>\n",
       "      <td>0.274534</td>\n",
       "      <td>0.725466</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>0.171583</td>\n",
       "      <td>1.080241</td>\n",
       "      <td>0.259627</td>\n",
       "      <td>0.740373</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>0.232525</td>\n",
       "      <td>0.954305</td>\n",
       "      <td>0.243478</td>\n",
       "      <td>0.756522</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>0.155711</td>\n",
       "      <td>0.969420</td>\n",
       "      <td>0.263354</td>\n",
       "      <td>0.736646</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>0.133733</td>\n",
       "      <td>0.837546</td>\n",
       "      <td>0.212422</td>\n",
       "      <td>0.787578</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>29</td>\n",
       "      <td>0.195431</td>\n",
       "      <td>0.987912</td>\n",
       "      <td>0.221118</td>\n",
       "      <td>0.778882</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>0.214681</td>\n",
       "      <td>0.939399</td>\n",
       "      <td>0.249689</td>\n",
       "      <td>0.750311</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>0.124636</td>\n",
       "      <td>0.754483</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>32</td>\n",
       "      <td>0.168588</td>\n",
       "      <td>0.819754</td>\n",
       "      <td>0.209938</td>\n",
       "      <td>0.790062</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>33</td>\n",
       "      <td>0.161661</td>\n",
       "      <td>0.791658</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>34</td>\n",
       "      <td>0.109011</td>\n",
       "      <td>0.782671</td>\n",
       "      <td>0.197516</td>\n",
       "      <td>0.802484</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>35</td>\n",
       "      <td>0.192774</td>\n",
       "      <td>0.789440</td>\n",
       "      <td>0.183851</td>\n",
       "      <td>0.816149</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>36</td>\n",
       "      <td>0.116204</td>\n",
       "      <td>0.625333</td>\n",
       "      <td>0.185093</td>\n",
       "      <td>0.814907</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37</td>\n",
       "      <td>0.115642</td>\n",
       "      <td>0.688804</td>\n",
       "      <td>0.178882</td>\n",
       "      <td>0.821118</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>38</td>\n",
       "      <td>0.163575</td>\n",
       "      <td>0.828804</td>\n",
       "      <td>0.183851</td>\n",
       "      <td>0.816149</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39</td>\n",
       "      <td>0.099515</td>\n",
       "      <td>0.641896</td>\n",
       "      <td>0.160248</td>\n",
       "      <td>0.839752</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>0.097081</td>\n",
       "      <td>0.602731</td>\n",
       "      <td>0.165217</td>\n",
       "      <td>0.834783</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41</td>\n",
       "      <td>0.121348</td>\n",
       "      <td>0.636313</td>\n",
       "      <td>0.156522</td>\n",
       "      <td>0.843478</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>42</td>\n",
       "      <td>0.073731</td>\n",
       "      <td>0.656654</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.857143</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>43</td>\n",
       "      <td>0.092405</td>\n",
       "      <td>0.526649</td>\n",
       "      <td>0.132919</td>\n",
       "      <td>0.867081</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>44</td>\n",
       "      <td>0.068598</td>\n",
       "      <td>0.442036</td>\n",
       "      <td>0.118012</td>\n",
       "      <td>0.881988</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45</td>\n",
       "      <td>0.086852</td>\n",
       "      <td>0.523134</td>\n",
       "      <td>0.139130</td>\n",
       "      <td>0.860870</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>46</td>\n",
       "      <td>0.078918</td>\n",
       "      <td>0.480999</td>\n",
       "      <td>0.134161</td>\n",
       "      <td>0.865839</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>47</td>\n",
       "      <td>0.065713</td>\n",
       "      <td>0.465827</td>\n",
       "      <td>0.124224</td>\n",
       "      <td>0.875776</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>48</td>\n",
       "      <td>0.075608</td>\n",
       "      <td>0.439734</td>\n",
       "      <td>0.108075</td>\n",
       "      <td>0.891925</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>49</td>\n",
       "      <td>0.076840</td>\n",
       "      <td>0.388335</td>\n",
       "      <td>0.113043</td>\n",
       "      <td>0.886957</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>0.054057</td>\n",
       "      <td>0.341099</td>\n",
       "      <td>0.095652</td>\n",
       "      <td>0.904348</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>51</td>\n",
       "      <td>0.034011</td>\n",
       "      <td>0.340049</td>\n",
       "      <td>0.101863</td>\n",
       "      <td>0.898137</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>52</td>\n",
       "      <td>0.057932</td>\n",
       "      <td>0.397950</td>\n",
       "      <td>0.100621</td>\n",
       "      <td>0.899379</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>53</td>\n",
       "      <td>0.059346</td>\n",
       "      <td>0.382785</td>\n",
       "      <td>0.111801</td>\n",
       "      <td>0.888199</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>54</td>\n",
       "      <td>0.051964</td>\n",
       "      <td>0.452730</td>\n",
       "      <td>0.113043</td>\n",
       "      <td>0.886957</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>55</td>\n",
       "      <td>0.038493</td>\n",
       "      <td>0.320593</td>\n",
       "      <td>0.093168</td>\n",
       "      <td>0.906832</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>56</td>\n",
       "      <td>0.062239</td>\n",
       "      <td>0.304702</td>\n",
       "      <td>0.090683</td>\n",
       "      <td>0.909317</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>57</td>\n",
       "      <td>0.033538</td>\n",
       "      <td>0.266728</td>\n",
       "      <td>0.079503</td>\n",
       "      <td>0.920497</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>58</td>\n",
       "      <td>0.035955</td>\n",
       "      <td>0.304774</td>\n",
       "      <td>0.098137</td>\n",
       "      <td>0.901863</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>59</td>\n",
       "      <td>0.036616</td>\n",
       "      <td>0.333432</td>\n",
       "      <td>0.093168</td>\n",
       "      <td>0.906832</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>60</td>\n",
       "      <td>0.044442</td>\n",
       "      <td>0.376899</td>\n",
       "      <td>0.106832</td>\n",
       "      <td>0.893168</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>61</td>\n",
       "      <td>0.038577</td>\n",
       "      <td>0.285362</td>\n",
       "      <td>0.090683</td>\n",
       "      <td>0.909317</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>62</td>\n",
       "      <td>0.022169</td>\n",
       "      <td>0.328253</td>\n",
       "      <td>0.085714</td>\n",
       "      <td>0.914286</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>63</td>\n",
       "      <td>0.022791</td>\n",
       "      <td>0.391093</td>\n",
       "      <td>0.103106</td>\n",
       "      <td>0.896894</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>64</td>\n",
       "      <td>0.042478</td>\n",
       "      <td>0.299289</td>\n",
       "      <td>0.081988</td>\n",
       "      <td>0.918012</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>65</td>\n",
       "      <td>0.035514</td>\n",
       "      <td>0.238172</td>\n",
       "      <td>0.073292</td>\n",
       "      <td>0.926708</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>66</td>\n",
       "      <td>0.022597</td>\n",
       "      <td>0.263564</td>\n",
       "      <td>0.086957</td>\n",
       "      <td>0.913043</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>67</td>\n",
       "      <td>0.020295</td>\n",
       "      <td>0.218213</td>\n",
       "      <td>0.065839</td>\n",
       "      <td>0.934161</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>68</td>\n",
       "      <td>0.035894</td>\n",
       "      <td>0.282178</td>\n",
       "      <td>0.074534</td>\n",
       "      <td>0.925466</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>69</td>\n",
       "      <td>0.017982</td>\n",
       "      <td>0.260687</td>\n",
       "      <td>0.072050</td>\n",
       "      <td>0.927950</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>70</td>\n",
       "      <td>0.014254</td>\n",
       "      <td>0.238768</td>\n",
       "      <td>0.074534</td>\n",
       "      <td>0.925466</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>71</td>\n",
       "      <td>0.018478</td>\n",
       "      <td>0.199174</td>\n",
       "      <td>0.064596</td>\n",
       "      <td>0.935404</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>72</td>\n",
       "      <td>0.020054</td>\n",
       "      <td>0.219791</td>\n",
       "      <td>0.059627</td>\n",
       "      <td>0.940373</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>73</td>\n",
       "      <td>0.012720</td>\n",
       "      <td>0.224474</td>\n",
       "      <td>0.065839</td>\n",
       "      <td>0.934161</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>74</td>\n",
       "      <td>0.010848</td>\n",
       "      <td>0.215075</td>\n",
       "      <td>0.060870</td>\n",
       "      <td>0.939130</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75</td>\n",
       "      <td>0.010696</td>\n",
       "      <td>0.227063</td>\n",
       "      <td>0.069565</td>\n",
       "      <td>0.930435</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>76</td>\n",
       "      <td>0.005756</td>\n",
       "      <td>0.189376</td>\n",
       "      <td>0.050932</td>\n",
       "      <td>0.949068</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>77</td>\n",
       "      <td>0.010287</td>\n",
       "      <td>0.185252</td>\n",
       "      <td>0.057143</td>\n",
       "      <td>0.942857</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78</td>\n",
       "      <td>0.015689</td>\n",
       "      <td>0.136085</td>\n",
       "      <td>0.039752</td>\n",
       "      <td>0.960248</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>79</td>\n",
       "      <td>0.005426</td>\n",
       "      <td>0.115787</td>\n",
       "      <td>0.039752</td>\n",
       "      <td>0.960248</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>80</td>\n",
       "      <td>0.009698</td>\n",
       "      <td>0.151767</td>\n",
       "      <td>0.048447</td>\n",
       "      <td>0.951553</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>81</td>\n",
       "      <td>0.005204</td>\n",
       "      <td>0.110152</td>\n",
       "      <td>0.034783</td>\n",
       "      <td>0.965217</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>82</td>\n",
       "      <td>0.004047</td>\n",
       "      <td>0.114986</td>\n",
       "      <td>0.034783</td>\n",
       "      <td>0.965217</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>83</td>\n",
       "      <td>0.004158</td>\n",
       "      <td>0.163543</td>\n",
       "      <td>0.053416</td>\n",
       "      <td>0.946584</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84</td>\n",
       "      <td>0.009039</td>\n",
       "      <td>0.138988</td>\n",
       "      <td>0.043478</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>85</td>\n",
       "      <td>0.006754</td>\n",
       "      <td>0.191037</td>\n",
       "      <td>0.052174</td>\n",
       "      <td>0.947826</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>86</td>\n",
       "      <td>0.003982</td>\n",
       "      <td>0.195469</td>\n",
       "      <td>0.049689</td>\n",
       "      <td>0.950311</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>87</td>\n",
       "      <td>0.003866</td>\n",
       "      <td>0.177618</td>\n",
       "      <td>0.049689</td>\n",
       "      <td>0.950311</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>88</td>\n",
       "      <td>0.004492</td>\n",
       "      <td>0.174466</td>\n",
       "      <td>0.043478</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>89</td>\n",
       "      <td>0.004034</td>\n",
       "      <td>0.152141</td>\n",
       "      <td>0.043478</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>90</td>\n",
       "      <td>0.001440</td>\n",
       "      <td>0.186529</td>\n",
       "      <td>0.050932</td>\n",
       "      <td>0.949068</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>91</td>\n",
       "      <td>0.003572</td>\n",
       "      <td>0.189599</td>\n",
       "      <td>0.050932</td>\n",
       "      <td>0.949068</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>92</td>\n",
       "      <td>0.001508</td>\n",
       "      <td>0.168098</td>\n",
       "      <td>0.047205</td>\n",
       "      <td>0.952795</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>93</td>\n",
       "      <td>0.001858</td>\n",
       "      <td>0.148741</td>\n",
       "      <td>0.034783</td>\n",
       "      <td>0.965217</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>94</td>\n",
       "      <td>0.000744</td>\n",
       "      <td>0.138702</td>\n",
       "      <td>0.040994</td>\n",
       "      <td>0.959006</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>95</td>\n",
       "      <td>0.002585</td>\n",
       "      <td>0.141475</td>\n",
       "      <td>0.039752</td>\n",
       "      <td>0.960248</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96</td>\n",
       "      <td>0.004510</td>\n",
       "      <td>0.135052</td>\n",
       "      <td>0.031056</td>\n",
       "      <td>0.968944</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>97</td>\n",
       "      <td>0.003152</td>\n",
       "      <td>0.136238</td>\n",
       "      <td>0.044720</td>\n",
       "      <td>0.955280</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>98</td>\n",
       "      <td>0.000858</td>\n",
       "      <td>0.136676</td>\n",
       "      <td>0.034783</td>\n",
       "      <td>0.965217</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>99</td>\n",
       "      <td>0.000519</td>\n",
       "      <td>0.143548</td>\n",
       "      <td>0.050932</td>\n",
       "      <td>0.949068</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>100</td>\n",
       "      <td>0.001485</td>\n",
       "      <td>0.119592</td>\n",
       "      <td>0.038509</td>\n",
       "      <td>0.961491</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>101</td>\n",
       "      <td>0.001062</td>\n",
       "      <td>0.121021</td>\n",
       "      <td>0.040994</td>\n",
       "      <td>0.959006</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>102</td>\n",
       "      <td>0.000551</td>\n",
       "      <td>0.099376</td>\n",
       "      <td>0.036025</td>\n",
       "      <td>0.963975</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>103</td>\n",
       "      <td>0.000613</td>\n",
       "      <td>0.107080</td>\n",
       "      <td>0.042236</td>\n",
       "      <td>0.957764</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104</td>\n",
       "      <td>0.000308</td>\n",
       "      <td>0.108137</td>\n",
       "      <td>0.036025</td>\n",
       "      <td>0.963975</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>105</td>\n",
       "      <td>0.000468</td>\n",
       "      <td>0.101996</td>\n",
       "      <td>0.039752</td>\n",
       "      <td>0.960248</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>106</td>\n",
       "      <td>0.001191</td>\n",
       "      <td>0.096008</td>\n",
       "      <td>0.031056</td>\n",
       "      <td>0.968944</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>107</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>0.108981</td>\n",
       "      <td>0.037267</td>\n",
       "      <td>0.962733</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>108</td>\n",
       "      <td>0.000911</td>\n",
       "      <td>0.101906</td>\n",
       "      <td>0.036025</td>\n",
       "      <td>0.963975</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>109</td>\n",
       "      <td>0.000985</td>\n",
       "      <td>0.106482</td>\n",
       "      <td>0.037267</td>\n",
       "      <td>0.962733</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>110</td>\n",
       "      <td>0.000230</td>\n",
       "      <td>0.105506</td>\n",
       "      <td>0.032298</td>\n",
       "      <td>0.967702</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>111</td>\n",
       "      <td>0.000306</td>\n",
       "      <td>0.099976</td>\n",
       "      <td>0.033540</td>\n",
       "      <td>0.966460</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>112</td>\n",
       "      <td>0.005453</td>\n",
       "      <td>0.099953</td>\n",
       "      <td>0.032298</td>\n",
       "      <td>0.967702</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>113</td>\n",
       "      <td>0.003114</td>\n",
       "      <td>0.102232</td>\n",
       "      <td>0.033540</td>\n",
       "      <td>0.966460</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>114</td>\n",
       "      <td>0.001258</td>\n",
       "      <td>0.102980</td>\n",
       "      <td>0.032298</td>\n",
       "      <td>0.967702</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>115</td>\n",
       "      <td>0.000452</td>\n",
       "      <td>0.102922</td>\n",
       "      <td>0.026087</td>\n",
       "      <td>0.973913</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>116</td>\n",
       "      <td>0.000287</td>\n",
       "      <td>0.099809</td>\n",
       "      <td>0.033540</td>\n",
       "      <td>0.966460</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>117</td>\n",
       "      <td>0.001599</td>\n",
       "      <td>0.097145</td>\n",
       "      <td>0.034783</td>\n",
       "      <td>0.965217</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>118</td>\n",
       "      <td>0.001247</td>\n",
       "      <td>0.103972</td>\n",
       "      <td>0.032298</td>\n",
       "      <td>0.967702</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>119</td>\n",
       "      <td>0.000188</td>\n",
       "      <td>0.105384</td>\n",
       "      <td>0.037267</td>\n",
       "      <td>0.962733</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fine_tune(120, freeze_epochs=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "035384d9-25d6-4bd3-8591-0214dcd2dfe2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Landmark2/pdo/Code/ALDAOps/aiking/book/Competitions/ultramnist\n"
     ]
    }
   ],
   "source": [
    "!pwd\n",
    "# learn.save(\"ultramnist_a0.pkl\")\n",
    "learn.export()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a672fd15-7556-4aa9-bc2d-8839a26ef3b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m      \u001b[0mlearn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mType:\u001b[0m           Learner\n",
       "\u001b[0;31mString form:\u001b[0m    <fastai.learner.Learner object at 0x152edfc161c0>\n",
       "\u001b[0;31mFile:\u001b[0m           /opt/anaconda/envs/aiking/lib/python3.9/site-packages/fastai/learner.py\n",
       "\u001b[0;31mDocstring:\u001b[0m      Group together a `model`, some `dls` and a `loss_func` to handle training\n",
       "\u001b[0;31mCall docstring:\u001b[0m Call `event_name` for all `Callback`s in `self.cbs`\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5a281150-4d05-49ca-bac2-b305a426eef2",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:aiking] *",
   "language": "python",
   "name": "conda-env-aiking-py"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}