siim_predict.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "incorporated-mobility",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import gc\n",
    "\n",
    "from PIL import Image\n",
    "from tqdm.auto import tqdm\n",
    "\n",
    "# pytorch lighting\n",
    "import pytorch_lightning as pl\n",
    "\n",
    "from src.config import *\n",
    "from src.dataset import *\n",
    "from src.models import *\n",
    "\n",
    "import warnings \n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "burning-lending",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dir = '/data/siim-covid19-detection/'\n",
    "test_csv_path = data_dir + 'sample_submission.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "antique-wilderness",
   "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>id</th>\n",
       "      <th>PredictionString</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00188a671292_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>004bd59708be_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>00508faccd39_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>006486aa80b2_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>00655178fdfc_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2472</th>\n",
       "      <td>46719b856de1_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2473</th>\n",
       "      <td>31c07523a69a_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2474</th>\n",
       "      <td>f77d7d1aebab_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2475</th>\n",
       "      <td>ccc5b63ca96d_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2476</th>\n",
       "      <td>5e8ac1fe2b82_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2477 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      id    PredictionString\n",
       "0     00188a671292_study  negative 1 0 0 1 1\n",
       "1     004bd59708be_study  negative 1 0 0 1 1\n",
       "2     00508faccd39_study  negative 1 0 0 1 1\n",
       "3     006486aa80b2_study  negative 1 0 0 1 1\n",
       "4     00655178fdfc_study  negative 1 0 0 1 1\n",
       "...                  ...                 ...\n",
       "2472  46719b856de1_image      none 1 0 0 1 1\n",
       "2473  31c07523a69a_image      none 1 0 0 1 1\n",
       "2474  f77d7d1aebab_image      none 1 0 0 1 1\n",
       "2475  ccc5b63ca96d_image      none 1 0 0 1 1\n",
       "2476  5e8ac1fe2b82_image      none 1 0 0 1 1\n",
       "\n",
       "[2477 rows x 2 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = pd.read_csv(test_csv_path)\n",
    "df_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "occupational-martial",
   "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>id</th>\n",
       "      <th>PredictionString</th>\n",
       "      <th>id_last_str</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00188a671292_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>004bd59708be_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>00508faccd39_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>006486aa80b2_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>00655178fdfc_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2472</th>\n",
       "      <td>46719b856de1_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>e</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2473</th>\n",
       "      <td>31c07523a69a_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>e</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2474</th>\n",
       "      <td>f77d7d1aebab_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>e</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2475</th>\n",
       "      <td>ccc5b63ca96d_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>e</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2476</th>\n",
       "      <td>5e8ac1fe2b82_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>e</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2477 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      id    PredictionString id_last_str\n",
       "0     00188a671292_study  negative 1 0 0 1 1           y\n",
       "1     004bd59708be_study  negative 1 0 0 1 1           y\n",
       "2     00508faccd39_study  negative 1 0 0 1 1           y\n",
       "3     006486aa80b2_study  negative 1 0 0 1 1           y\n",
       "4     00655178fdfc_study  negative 1 0 0 1 1           y\n",
       "...                  ...                 ...         ...\n",
       "2472  46719b856de1_image      none 1 0 0 1 1           e\n",
       "2473  31c07523a69a_image      none 1 0 0 1 1           e\n",
       "2474  f77d7d1aebab_image      none 1 0 0 1 1           e\n",
       "2475  ccc5b63ca96d_image      none 1 0 0 1 1           e\n",
       "2476  5e8ac1fe2b82_image      none 1 0 0 1 1           e\n",
       "\n",
       "[2477 rows x 3 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "id_laststr_list  = []\n",
    "for i in range(df_test.shape[0]):\n",
    "    id_laststr_list.append(df_test.loc[i,'id'][-1])\n",
    "df_test['id_last_str'] = id_laststr_list\n",
    "df_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "chronic-wales",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1214"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "study_len = df_test[df_test['id_last_str'] == 'y'].shape[0]\n",
    "study_len"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "experimental-citation",
   "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>id</th>\n",
       "      <th>PredictionString</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00188a671292_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>004bd59708be_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>00508faccd39_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>006486aa80b2_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>00655178fdfc_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2472</th>\n",
       "      <td>46719b856de1_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2473</th>\n",
       "      <td>31c07523a69a_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2474</th>\n",
       "      <td>f77d7d1aebab_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2475</th>\n",
       "      <td>ccc5b63ca96d_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2476</th>\n",
       "      <td>5e8ac1fe2b82_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2477 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      id    PredictionString\n",
       "0     00188a671292_study  negative 1 0 0 1 1\n",
       "1     004bd59708be_study  negative 1 0 0 1 1\n",
       "2     00508faccd39_study  negative 1 0 0 1 1\n",
       "3     006486aa80b2_study  negative 1 0 0 1 1\n",
       "4     00655178fdfc_study  negative 1 0 0 1 1\n",
       "...                  ...                 ...\n",
       "2472  46719b856de1_image      none 1 0 0 1 1\n",
       "2473  31c07523a69a_image      none 1 0 0 1 1\n",
       "2474  f77d7d1aebab_image      none 1 0 0 1 1\n",
       "2475  ccc5b63ca96d_image      none 1 0 0 1 1\n",
       "2476  5e8ac1fe2b82_image      none 1 0 0 1 1\n",
       "\n",
       "[2477 rows x 2 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df = pd.read_csv(test_csv_path)\n",
    "test_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "underlying-bench",
   "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>id</th>\n",
       "      <th>PredictionString</th>\n",
       "      <th>negative</th>\n",
       "      <th>typical</th>\n",
       "      <th>indeterminate</th>\n",
       "      <th>atypical</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00188a671292_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>004bd59708be_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>00508faccd39_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>006486aa80b2_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>00655178fdfc_study</td>\n",
       "      <td>negative 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2472</th>\n",
       "      <td>46719b856de1_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2473</th>\n",
       "      <td>31c07523a69a_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2474</th>\n",
       "      <td>f77d7d1aebab_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2475</th>\n",
       "      <td>ccc5b63ca96d_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2476</th>\n",
       "      <td>5e8ac1fe2b82_image</td>\n",
       "      <td>none 1 0 0 1 1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2477 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      id    PredictionString  negative  typical  \\\n",
       "0     00188a671292_study  negative 1 0 0 1 1         0        0   \n",
       "1     004bd59708be_study  negative 1 0 0 1 1         0        0   \n",
       "2     00508faccd39_study  negative 1 0 0 1 1         0        0   \n",
       "3     006486aa80b2_study  negative 1 0 0 1 1         0        0   \n",
       "4     00655178fdfc_study  negative 1 0 0 1 1         0        0   \n",
       "...                  ...                 ...       ...      ...   \n",
       "2472  46719b856de1_image      none 1 0 0 1 1         0        0   \n",
       "2473  31c07523a69a_image      none 1 0 0 1 1         0        0   \n",
       "2474  f77d7d1aebab_image      none 1 0 0 1 1         0        0   \n",
       "2475  ccc5b63ca96d_image      none 1 0 0 1 1         0        0   \n",
       "2476  5e8ac1fe2b82_image      none 1 0 0 1 1         0        0   \n",
       "\n",
       "      indeterminate  atypical  \n",
       "0                 0         0  \n",
       "1                 0         0  \n",
       "2                 0         0  \n",
       "3                 0         0  \n",
       "4                 0         0  \n",
       "...             ...       ...  \n",
       "2472              0         0  \n",
       "2473              0         0  \n",
       "2474              0         0  \n",
       "2475              0         0  \n",
       "2476              0         0  \n",
       "\n",
       "[2477 rows x 6 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df['negative'] = 0\n",
    "test_df['typical'] = 0\n",
    "test_df['indeterminate'] = 0\n",
    "test_df['atypical'] = 0\n",
    "test_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "patient-victoria",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['negative', 'typical', 'indeterminate', 'atypical'], dtype='object')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label_cols = test_df.columns[2:]\n",
    "label_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "equipped-regulation",
   "metadata": {},
   "outputs": [],
   "source": [
    "CFG = Config"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "everyday-appliance",
   "metadata": {},
   "outputs": [],
   "source": [
    "def to_tensor(x, **kwargs):\n",
    "    if x.ndim==2 : \n",
    "        x = np.expand_dims(x,2)\n",
    "    x = np.transpose(x,(2,0,1)).astype('float32') / 255.\n",
    "    \n",
    "    x = torch.from_numpy(x)\n",
    "    return x\n",
    "\n",
    "def get_preprocessing():\n",
    "    \"\"\"Construct preprocessing transform\n",
    "    \n",
    "    Args:\n",
    "        preprocessing_fn (callbale): data normalization function \n",
    "            (can be specific for each pretrained neural network)\n",
    "    Return:\n",
    "        transform: albumentations.Compose\n",
    "    \n",
    "    \"\"\"\n",
    "    \n",
    "    _transform = [\n",
    "#         A.Lambda(image=preprocessing_fn),\n",
    "        A.Lambda(image=to_tensor, mask=to_tensor),\n",
    "    ]\n",
    "    return A.Compose(_transform)\n",
    "\n",
    "\n",
    "def get_val_transforms(CFG):\n",
    "    return A.Compose([\n",
    "            A.Resize(CFG.image_size, CFG.image_size),\n",
    "#             A.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225], max_pixel_value=255.0, p=1.0),\n",
    "#             ToTensorV2(),\n",
    "        ],p=1.0)\n",
    "\n",
    "class SIIMTestDataset(Dataset):\n",
    "    def __init__(self, df, transforms=None, preprocessing=None):\n",
    "        super().__init__()\n",
    "        self.df = df\n",
    "        self.transforms = transforms\n",
    "        self.preprocessing = preprocessing\n",
    "        self.length = len(df)\n",
    "        \n",
    "    def __len__(self):\n",
    "        return self.length\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        d = self.df.iloc[index]\n",
    "        split = d.id.split('_')[-1]\n",
    "        \n",
    "        if split == 'study':\n",
    "            image_path =  '/home/chen/ai-competition/siim-covid19-detection/test/study/%s.png' % (d.id)\n",
    "        else:\n",
    "            image_path =  '/home/chen/ai-competition/siim-covid19-detection/test/image/%s.png' % (d.id)\n",
    "        image = cv2.imread(image_path)\n",
    "        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
    "        if image is None:\n",
    "            raise FileNotFoundError(image_path)\n",
    "        \n",
    "        # apply augmentations\n",
    "        if self.transforms:\n",
    "            image = self.transforms(image=image)['image']\n",
    "        else:\n",
    "            image = torch.from_numpy(image)\n",
    "            \n",
    "        # apply preprocessing\n",
    "        if self.preprocessing:\n",
    "            sample = self.preprocessing(image=image)\n",
    "            image = sample['image']\n",
    "    \n",
    "        return image\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "sensitive-native",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_dataset = SIIMTestDataset(test_df, transforms=get_val_transforms(CFG), preprocessing=get_preprocessing())\n",
    "test_loader = DataLoader(test_dataset, 32, num_workers=4, shuffle=False) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "proud-paper",
   "metadata": {},
   "outputs": [],
   "source": [
    "class SIIMPLModel(pl.LightningModule):\n",
    "    def __init__(self):\n",
    "        super(SIIMPLModel,self).__init__()\n",
    "        self.model = SIIMMaskNet(CFG.model_name, CFG.num_classes, pretrained=False)\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.model(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "accompanied-cigarette",
   "metadata": {},
   "outputs": [],
   "source": [
    "def do_predict(model, test_loader, tta=['']):\n",
    "    print(f'tta is {tta}')\n",
    "    test_probability = []\n",
    "    test_num = 0\n",
    "    tk0 = tqdm(enumerate(test_loader), total=len(test_loader))\n",
    "    for t, (image) in tk0:\n",
    "        batch_size = image.size(0)\n",
    "        image = image.to(device)\n",
    "        #<todo> TTA\n",
    "        model.eval()\n",
    "        with torch.no_grad():\n",
    "            probability = []\n",
    "            logit, mask = model(image)\n",
    "            probability.append(F.softmax(logit,-1))\n",
    "\n",
    "            if 'flip' in tta:\n",
    "                logit, mask = model(torch.flip(image,dims=(3,)))\n",
    "                probability.append(F.softmax(logit,-1))\n",
    "\n",
    "            if 'scale' in tta:\n",
    "                # size=None, scale_factor=None, mode='nearest', align_corners=None, recompute_scale_factor=None):\n",
    "                logit, mask = model(F.interpolate(image, scale_factor=1.25, mode='bilinear', align_corners=False))\n",
    "                probability.append(F.softmax(logit,-1))\n",
    "\n",
    "            #--------------\n",
    "            probability = torch.stack(probability,0).mean(0)\n",
    "\n",
    "        test_num += batch_size\n",
    "        test_probability.append(probability.data.cpu().numpy())\n",
    "    assert(test_num == len(test_loader.dataset))\n",
    "\n",
    "    probability = np.concatenate(test_probability)\n",
    "    return probability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "lonely-beauty",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ====================================================\n",
    "# Helper functions\n",
    "# ====================================================\n",
    "def inference(model, states, test_loader, device, ttas):\n",
    "    probs = 0\n",
    "    for (state,tta) in zip(states,ttas):\n",
    "        model = model.load_from_checkpoint(state)\n",
    "        model.to(device)\n",
    "        probability = do_predict(model, test_loader, tta)\n",
    "        probs += probability**0.5\n",
    "    probs = probs/len(states) \n",
    "    return probs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "interpreted-density",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tta is ['flip']\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d2bded1ad4084fc3bb7a9a42fb19f848",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value=''), FloatProgress(value=0.0, max=78.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-15-97e54f1dadcd>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     28\u001b[0m     \u001b[0;34m[\u001b[0m\u001b[0;34m'flip'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     29\u001b[0m     ]\n\u001b[0;32m---> 30\u001b[0;31m \u001b[0mpredictions\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0minference\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpl_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdevice\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfold_ttas\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m<ipython-input-14-a4aa748c778d>\u001b[0m in \u001b[0;36minference\u001b[0;34m(model, states, test_loader, device, ttas)\u001b[0m\n\u001b[1;32m      7\u001b[0m         \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_from_checkpoint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstate\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m         \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m         \u001b[0mprobability\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdo_predict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtta\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     10\u001b[0m         \u001b[0mprobs\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mprobability\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0;36m0.5\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     11\u001b[0m     \u001b[0mprobs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mprobs\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m<ipython-input-13-d20fc1cfe1b1>\u001b[0m in \u001b[0;36mdo_predict\u001b[0;34m(model, test_loader, tta)\u001b[0m\n\u001b[1;32m     15\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     16\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0;34m'flip'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtta\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 17\u001b[0;31m                 \u001b[0mlogit\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmask\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimage\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mdims\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     18\u001b[0m                 \u001b[0mprobability\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msoftmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlogit\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     19\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\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;32m<ipython-input-12-6789934759c7>\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\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;32m----> 7\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\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;32m~/projects/pl-siim-covid19-detection/src/models.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    187\u001b[0m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mblocks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    188\u001b[0m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mblocks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 189\u001b[0;31m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mblocks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    190\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    191\u001b[0m             \u001b[0;31m####----\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\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;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/container.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    115\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    116\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 117\u001b[0;31m             \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    118\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    119\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\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;32m~/ai-competition/pytorch-image-models-master-210703/timm/models/efficientnet_blocks.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    193\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    194\u001b[0m         \u001b[0;31m# Squeeze-and-excitation\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 195\u001b[0;31m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mse\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    196\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    197\u001b[0m         \u001b[0;31m# Point-wise linear projection\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\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;32m~/ai-competition/pytorch-image-models-master-210703/timm/models/efficientnet_blocks.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m     44\u001b[0m         \u001b[0mx_se\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv_reduce\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_se\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     45\u001b[0m         \u001b[0mx_se\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mact1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_se\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 46\u001b[0;31m         \u001b[0mx_se\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv_expand\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_se\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     47\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_se\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     48\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\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;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/conv.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    421\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    422\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 423\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_conv_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mweight\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    424\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    425\u001b[0m \u001b[0;32mclass\u001b[0m \u001b[0mConv3d\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_ConvNd\u001b[0m\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;32m/data/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/conv.py\u001b[0m in \u001b[0;36m_conv_forward\u001b[0;34m(self, input, weight)\u001b[0m\n\u001b[1;32m    418\u001b[0m                             _pair(0), self.dilation, self.groups)\n\u001b[1;32m    419\u001b[0m         return F.conv2d(input, weight, self.bias, self.stride,\n\u001b[0;32m--> 420\u001b[0;31m                         self.padding, self.dilation, self.groups)\n\u001b[0m\u001b[1;32m    421\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    422\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# ====================================================\n",
    "# inference\n",
    "# ====================================================\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "    \n",
    "### init pl model\n",
    "pl_model = SIIMPLModel()\n",
    "\n",
    "#### drop duplicate images and save one 6227\n",
    "models_path = f'/data/output/pl-siim-covid19-study-classification/\\\n",
    "{CFG.model_name}_folds_cutout_{CFG.image_size}_30e_Adam_GradualWarmupSchedulerV2_\\\n",
    "LabelSmoothingBinaryCrossEntropy_ls0.0_lovasz0.0_v2_dropdup'\n",
    "\n",
    "### effnet-b5 best cv=0.3877336, lb=0.535\n",
    "states = [\n",
    "    f'{models_path}/{CFG.model_name}/fold_0/epoch=17_mAP=0.387950.ckpt',\n",
    "    f'{models_path}/{CFG.model_name}/fold_1/epoch=17_mAP=0.401342.ckpt',\n",
    "    f'{models_path}/{CFG.model_name}/fold_2/epoch=17_mAP=0.382397.ckpt',\n",
    "    f'{models_path}/{CFG.model_name}/fold_3/epoch=20_mAP=0.370457.ckpt',\n",
    "    f'{models_path}/{CFG.model_name}/fold_4/epoch=18_mAP=0.396522.ckpt',\n",
    "]\n",
    "\n",
    "fold_ttas = [\n",
    "    ['flip'],\n",
    "    ['flip'],\n",
    "    ['flip'],\n",
    "    ['flip'],\n",
    "    ['flip'],\n",
    "    ]\n",
    "predictions = inference(pl_model, states, test_loader, device, fold_ttas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "naked-friend",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "finnish-province",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df[label_cols] = predictions\n",
    "test_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "constant-decline",
   "metadata": {},
   "outputs": [],
   "source": [
    "# test_df.to_csv(f'./predicts/{CFG.model_name}_folds_{CFG.image_size}_study_siim_test_ricord_pseudo_v3_flip.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "directed-situation",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "american-adjustment",
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}