Beaf_classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "PGH08Ycnsrdv",
    "outputId": "e1efed64-0daa-44f9-c9b9-2b534945087d"
   },
   "outputs": [],
   "source": [
    "!git clone https://github.com/AhMuGeoNa/2022-AI-Competition.git"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "onRtrMRdtudK",
    "outputId": "38c147ae-c0a6-489d-b0ef-71de172425e7"
   },
   "outputs": [],
   "source": [
    "!cp -r 2022-AI-Competition/data data\n",
    "!mkdir data/csv\n",
    "!mkdir data/zips\n",
    "!cp -r train.zip data/zips/train.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!mv data/train.csv data/csv/train.csv\n",
    "!mv data/test.csv data/csv/test.csv\n",
    "!mv data/valid.csv data/csv/valid.csv\n",
    "\n",
    "!unzip data/zips/train.zip -d data/images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "QnsYtadwt2oQ"
   },
   "outputs": [],
   "source": [
    "## 데이터 처리에 필요한 패키지 로드\n",
    "from imutils.paths import list_images\n",
    "import matplotlib.pyplot as plt\n",
    "from collections import Counter\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import cv2, os\n",
    "import shutil"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XSQYCSwVwS9t"
   },
   "source": [
    "# **1.데이터셋 전처리**\n",
    "- 전체 이미지 데이터 셋 10,000장\n",
    "- 학습 : 시험 : 검증 = 6 : 2 : 2로 분할\n",
    "- 이미지 중에서 어두운 이미지가 존재하여 이미지 평탄화 진행"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "jtjW0EjXvccC"
   },
   "outputs": [],
   "source": [
    "images_path = 'data/images/train'\n",
    "csv_path    = 'data/csv'\n",
    "\n",
    "## csv 파일에 존재하는 정보들을 기반으로하여 라벨별로 데이터 정리하는 함수\n",
    "def data_organizer(dtype = 'train'):\n",
    "  df = pd.read_csv(f'{csv_path}/{dtype}.csv')\n",
    "\n",
    "  for image_path, grade in zip(df['imname'], df['grade']):\n",
    "    os.makedirs(f'data/{dtype}/{grade}', exist_ok = True)\n",
    "    shutil.copy(f'{images_path}/{grade}/{image_path}', f'data/{dtype}/{grade}/{image_path}')\n",
    "\n",
    "data_organizer()\n",
    "data_organizer(dtype = 'test')\n",
    "data_organizer(dtype = 'valid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 269
    },
    "id": "sbObNy2Dw43b",
    "outputId": "baa58b57-26b6-4791-cb3c-21c251a23967"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## 데이터 분포 확인하는 함수\n",
    "def data_distribution(data_path):\n",
    "  ## data_path에 있는 이미지 목록을 리스트로 불러오는 부분.\n",
    "  image_paths = sorted(list_images(data_path))\n",
    "\n",
    "  ## 이미지 경로에서 '/'을 기준으로 나눈것 중 마지막에서 2번째 값만 가져옴. (라벨값만 가져옴.)\n",
    "  ## (e.g.) data/train/1++/cow_1++_1002.jpg => 1++\n",
    "  image_paths = [image_path.split(os.path.sep)[-2] for image_path in image_paths]\n",
    "\n",
    "  ## 라벨별 개수를 반환해주는 부분 (딕셔너리 형태로 만들어짐)\n",
    "  label_count = Counter(image_paths)\n",
    "  label_count = dict(sorted(label_count.items()))\n",
    "\n",
    "  ## 레이블 별 데이터 분포를 시각화 시키는 코드\n",
    "  plt.bar(label_count.keys(), label_count.values())\n",
    "  plt.ylim(0, 2000)\n",
    "\n",
    "  for grade, num in label_count.items():\n",
    "      plt.text(\n",
    "                  grade, num, num,\n",
    "                  fontsize = 11, color = 'black',\n",
    "                  horizontalalignment = 'center',\n",
    "                  verticalalignment = 'bottom'\n",
    "              )\n",
    "  plt.show()\n",
    "  return label_count\n",
    "## 3등급 이미지가 다른 이미지들에 비해 수가 적은 것을 보임.\n",
    "## 일단 그냥 한 번 돌려보고 성능이 제대로 나오지 않을 경우에 불균형 해소후 학습\n",
    "train_label_count = data_distribution('data/train')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "X8FrQ_ESDlqq",
    "outputId": "662cc770-1527-46c9-df05-4a584182912a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'1++': 0, '1+': 1, '1': 2, '2': 3, '3': 4}\n"
     ]
    }
   ],
   "source": [
    "## 라벨을 인덱스 값으로 변경\n",
    "label2idx = {'1++' : 0, '1+' : 1, '1' :2, '2' :3, '3': 4}\n",
    "print(label2idx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "id": "xXpYJn9L00IC"
   },
   "outputs": [],
   "source": [
    "## 이미지 전처리 해주는 함수\n",
    "def preprocessing(image):\n",
    "    image = cv2.cvtColor(image, cv2.COLOR_BGR2YCrCb)\n",
    "    ycrcb_planes = list(cv2.split(image))\n",
    "\n",
    "\n",
    "    # 밝기 성분에 대해서만 히스토그램 평활화 수행\n",
    "    ycrcb_planes[0] = cv2.equalizeHist(ycrcb_planes[0])\n",
    "\n",
    "    image = cv2.merge(ycrcb_planes)\n",
    "    image = cv2.cvtColor(image, cv2.COLOR_YCrCb2RGB)\n",
    "\n",
    "    return image\n",
    "\n",
    "datasets = {}\n",
    "## 전처리한 이미지를 넘파이 배열로 만들어주는 함수\n",
    "def build_dataset(data_path, dtype = 'train'):\n",
    "  image_paths = sorted(list_images(data_path))\n",
    "\n",
    "  images, labels = [], []\n",
    "  for image_path in image_paths:\n",
    "\n",
    "    image = cv2.imread(image_path)\n",
    "    image = preprocessing(image)\n",
    "    \n",
    "    images.append(image)\n",
    "    labels.append(label2idx[image_path.split(os.path.sep)[-2]])\n",
    "\n",
    "  datasets[dtype] = [np.array(images), np.array(labels)]\n",
    "  return datasets\n",
    "\n",
    "datasets = build_dataset('data/train')\n",
    "datasets = build_dataset('data/test', dtype = 'test')\n",
    "datasets = build_dataset('data/valid', dtype = 'valid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 266
    },
    "id": "0fHcoXYT9_CY",
    "outputId": "38b1b6c4-b9a1-4595-df17-44d801aaa9e3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f9925500280>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## 원본 이미지\n",
    "image = cv2.imread('data/train/1++/cow_1++_1074.jpg')\n",
    "plt.axis(False)\n",
    "plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 266
    },
    "id": "MoqZYSBp05NZ",
    "outputId": "a0eed596-aaca-44f0-af4d-ae247b005f0b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f99256a3f40>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## 전처리한 이미지\n",
    "image = preprocessing(image)\n",
    "\n",
    "plt.axis(False)\n",
    "plt.imshow(image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UeWUsrGaXcmL"
   },
   "source": [
    "# **2. 학습 진행하기**\n",
    "## **2-1. 데이터 셋 구성하기**\n",
    "- opencv로 구성한 데이터 셋을 pytorch 데이터 셋으로 변경\n",
    "\n",
    "#### **데이터 셋 픽셀 평균, 표준편차 값 구해주는 함수 참조**\n",
    "- Inhovative AI | 3. (1) 이미지 전처리 - augmentation, normalization | [[블로그 주소]]('https://inhovation97.tistory.com/37')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "TC419I6oHaGI",
    "outputId": "40d59af7-6d1a-426c-8fb9-8e31223cd978"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mWARNING: The directory '/home/jovyan/.cache/pip' or its parent directory is not owned or is not writable by the current user. The cache has been disabled. Check the permissions and owner of that directory. If executing pip with sudo, you should use sudo's -H flag.\u001b[0m\n",
      "Requirement already satisfied: timm in /opt/conda/lib/python3.9/site-packages (0.5.4)\n",
      "Requirement already satisfied: torchvision in /opt/conda/lib/python3.9/site-packages (from timm) (0.12.0)\n",
      "Requirement already satisfied: torch>=1.4 in /opt/conda/lib/python3.9/site-packages (from timm) (1.11.0)\n",
      "Requirement already satisfied: typing-extensions in /opt/conda/lib/python3.9/site-packages (from torch>=1.4->timm) (3.10.0.2)\n",
      "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /opt/conda/lib/python3.9/site-packages (from torchvision->timm) (9.1.1)\n",
      "Requirement already satisfied: requests in /opt/conda/lib/python3.9/site-packages (from torchvision->timm) (2.26.0)\n",
      "Requirement already satisfied: numpy in /opt/conda/lib/python3.9/site-packages (from torchvision->timm) (1.20.3)\n",
      "Requirement already satisfied: charset-normalizer~=2.0.0 in /opt/conda/lib/python3.9/site-packages (from requests->torchvision->timm) (2.0.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.9/site-packages (from requests->torchvision->timm) (2021.10.8)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.9/site-packages (from requests->torchvision->timm) (1.26.7)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.9/site-packages (from requests->torchvision->timm) (3.1)\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "## 사전 학습된 모델을 불러오는 패키지\n",
    "!pip install timm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0iu-iih4Y8g7",
    "outputId": "75bbece9-0776-441e-cd63-7678ca38e9a0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[33mWARNING: The directory '/home/jovyan/.cache/pip' or its parent directory is not owned or is not writable by the current user. The cache has been disabled. Check the permissions and owner of that directory. If executing pip with sudo, you should use sudo's -H flag.\u001b[0m\n",
      "Collecting git+https://github.com/katsura-jp/pytorch-cosine-annealing-with-warmup\n",
      "  Cloning https://github.com/katsura-jp/pytorch-cosine-annealing-with-warmup to /tmp/pip-req-build-hkmjkvne\n",
      "  Running command git clone --filter=blob:none -q https://github.com/katsura-jp/pytorch-cosine-annealing-with-warmup /tmp/pip-req-build-hkmjkvne\n",
      "  Resolved https://github.com/katsura-jp/pytorch-cosine-annealing-with-warmup to commit 2958117072f214dce743f918d6a2253f682ab65d\n",
      "  Preparing metadata (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25hBuilding wheels for collected packages: cosine-annealing-warmup\n",
      "  Building wheel for cosine-annealing-warmup (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25h  Created wheel for cosine-annealing-warmup: filename=cosine_annealing_warmup-2.0-py3-none-any.whl size=3334 sha256=3aaae2473a75ba1961610c101d088a93be876f2cf9106523269662b2b3f8eb31\n",
      "  Stored in directory: /tmp/pip-ephem-wheel-cache-lzv28brd/wheels/a6/ce/7d/6fe229886a381b3af560c1d75942eff933a959781ff693b89b\n",
      "Successfully built cosine-annealing-warmup\n",
      "Installing collected packages: cosine-annealing-warmup\n",
      "Successfully installed cosine-annealing-warmup-2.0\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "!pip install 'git+https://github.com/katsura-jp/pytorch-cosine-annealing-with-warmup'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "id": "s2BS684tHS6B"
   },
   "outputs": [],
   "source": [
    "## 신경망 구성에 필요한 패키지 로드\n",
    "from torch.utils.data import WeightedRandomSampler, DataLoader, Dataset\n",
    "from torch.optim import Adam, AdamW\n",
    "from torchvision import transforms\n",
    "import torch.nn.functional as F\n",
    "from torchvision import utils\n",
    "import torch.nn as nn\n",
    "import torch\n",
    "import timm\n",
    "import gc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 330,
   "metadata": {
    "id": "sg7U1JmaHvwq"
   },
   "outputs": [],
   "source": [
    "NUM_CLASSES     = 5\n",
    "SEED            = 99\n",
    "DEVICE          = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
    "\n",
    "NUM_WORKERS     = 1\n",
    "PIN_MEMORY      = True\n",
    "\n",
    "DROPOUT_RATE    = 0.5\n",
    "EPOCHS          = 100\n",
    "BATCH_SIZE      = 32\n",
    "LR              = 1e-3\n",
    "CLIP_GRAD       = 10\n",
    "PATIENCE        = 15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {
    "id": "R_3J8IUbko5h"
   },
   "outputs": [],
   "source": [
    "## 데이터셋 픽셀 평균, 표준편차 값 구해주는 함수\n",
    "def get_mean_std(dataset):\n",
    "  to_tensor = transforms.Compose([\n",
    "                                  transforms.ToPILImage(),\n",
    "                                  transforms.ToTensor()\n",
    "                                ])\n",
    "  dataset = [to_tensor(data) for data in dataset]\n",
    "  \n",
    "  meanRGB = [np.mean(image.numpy(), axis=(1, 2)) for image in dataset]\n",
    "  stdRGB  = [np.std(image.numpy(), axis=(1, 2)) for image in dataset]\n",
    "\n",
    "  meanR = np.mean([m[0] for m in meanRGB])\n",
    "  meanG = np.mean([m[1] for m in meanRGB])\n",
    "  meanB = np.mean([m[2] for m in meanRGB])\n",
    "\n",
    "  stdR = np.mean([s[0] for s in stdRGB])\n",
    "  stdG = np.mean([s[1] for s in stdRGB])\n",
    "  stdB = np.mean([s[2] for s in stdRGB])\n",
    "\n",
    "  return {'mean' : (meanR, meanG, meanB), 'std' : (stdR, stdG, stdB)}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 333,
   "metadata": {},
   "outputs": [],
   "source": [
    "## 커스텀 데이터셋 생성\n",
    "class BeafData(Dataset):\n",
    "\n",
    "  def __init__(self, images, labels, mean_n_std, dtype = 'train'):\n",
    "\n",
    "    self.images = images\n",
    "    self.labels = labels\n",
    "    self.dtype  = dtype\n",
    "    \n",
    "    self.transform = {}\n",
    "    self.transform['train'] = transforms.Compose([\n",
    "                                    transforms.ToPILImage(),\n",
    "                                    transforms.Resize((240, 240)),\n",
    "                                    transforms.RandomHorizontalFlip(),\n",
    "                                    transforms.RandomVerticalFlip(),\n",
    "                                    transforms.ToTensor(),\n",
    "                                    transforms.Normalize(mean_n_std['mean'], mean_n_std['std'])             \n",
    "                              ])\n",
    "    self.transform['test'] = transforms.Compose([\n",
    "                                transforms.ToPILImage(),\n",
    "                                transforms.Resize((240, 240)),\n",
    "                                transforms.ToTensor(),\n",
    "                                transforms.Normalize(mean_n_std['mean'], mean_n_std['std'])          \n",
    "                          ])\n",
    "\n",
    "    self.transform['valid'] = transforms.Compose([\n",
    "                            transforms.ToPILImage(),\n",
    "                            transforms.Resize((240, 240)),\n",
    "                            transforms.ToTensor(),\n",
    "                            transforms.Normalize(mean_n_std['mean'], mean_n_std['std'])          \n",
    "                      ])\n",
    "\n",
    "    self.num_datas = self.labels.shape[0]\n",
    "\n",
    "  ## 이미지, 라벨 데이터를 튜플형태로 변환\n",
    "  def __getitem__(self, index):\n",
    "    image = self.images[index]\n",
    "\n",
    "    image = self.transform[self.dtype](image)\n",
    "    return image, self.labels[index]\n",
    "\n",
    "  def __len__(self):\n",
    "    return self.num_datas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 334,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'mean': (0.43284187, 0.31426674, 0.30875233), 'std': (0.3776003, 0.3116054, 0.31011364)}\n"
     ]
    }
   ],
   "source": [
    "train_images, train_labels = datasets['train']\n",
    "train_mean_n_std = get_mean_std(train_images)\n",
    "print(train_mean_n_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 335,
   "metadata": {
    "id": "XQ3yqLOb1fSX"
   },
   "outputs": [],
   "source": [
    "## 데이터 불균형 문제로 로스가 안떨이지는 것으로 가정하고, 학습 진행해보기\n",
    "class_counts  = [count for count in train_label_count.values()]\n",
    "total_samples = sum(class_counts)\n",
    "\n",
    "## labels        | 학습 데이터 셋의 라벨인덱스 값들 [0, 0, 0, 0, ..., 4, 4, 4, 4]\n",
    "## class_weights | 라벨별 가중치 설정 (수가 적은 라벨에 더 많은 가중치 부여)\n",
    "labels = [lb for lb in train_labels.data]\n",
    "class_weights = [total_samples / class_count for class_count in class_counts]\n",
    "weights = [class_weights[lb_idx] for lb_idx in labels]\n",
    "\n",
    "sampler = WeightedRandomSampler(torch.DoubleTensor(weights), int(total_samples))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 336,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 324
    },
    "id": "RFah6SYCnCR-",
    "outputId": "5dfa28ec-5e7e-4a99-c3db-25798f99bc07"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([3, 240, 240])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f98ca3f2760>"
      ]
     },
     "execution_count": 336,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_dataset = BeafData(train_images, train_labels, mean_n_std = train_mean_n_std)\n",
    "train_loader  = DataLoader(train_dataset, batch_size = BATCH_SIZE,\n",
    "                           sampler = sampler, pin_memory = PIN_MEMORY, num_workers = NUM_WORKERS)\n",
    "\n",
    "data_iter = iter(train_loader)\n",
    "\n",
    "images, labels = next(data_iter)\n",
    "image = images[0]\n",
    "\n",
    "print(image.shape)\n",
    "plt.imshow(np.transpose(image, (1, 2, 0)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 337,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'mean': (0.43203428, 0.31337285, 0.30804688), 'std': (0.37785816, 0.3113508, 0.30983475)}\n"
     ]
    }
   ],
   "source": [
    "valid_images, valid_labels = datasets['valid']\n",
    "valid_mean_n_std = get_mean_std(valid_images)\n",
    "print(valid_mean_n_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 338,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 324
    },
    "id": "Bbk2j5pOCpaw",
    "outputId": "f8ec2af3-ae8d-4a00-e762-3d5c8dabf716"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([3, 240, 240])\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f98ca7b6df0>"
      ]
     },
     "execution_count": 338,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "valid_dataset = BeafData(valid_images, valid_labels, mean_n_std = valid_mean_n_std, dtype = 'valid')\n",
    "valid_loader  = DataLoader(valid_dataset, batch_size = BATCH_SIZE, shuffle = False)\n",
    "\n",
    "data_iter = iter(valid_loader)\n",
    "images, labels = next(data_iter)\n",
    "image = images[0]\n",
    "\n",
    "print(image.shape)\n",
    "plt.imshow(np.transpose(image, (1, 2, 0)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "W12w7e4s4wLy"
   },
   "source": [
    "## **2-2. 신경망 구성하기**\n",
    "- timm을 이용한 사전 학습모델 사용하여 신경망 구성\n",
    "- 이미지 데이터가 248 x 200 이미지 이므로,   \n",
    "  사전 학습 모델의 입력 사이즈가 240 x 240인 것으로 테스트 해보기  \n",
    "\n",
    "#### **Timm 사용법, 학습 과정 참조**\n",
    "- 프로그래머스 공식 블로그 | [[블로그 주소]](https://prgms.tistory.com/75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "OXo9V7Lf7ZCC",
    "outputId": "a21ff22c-4578-4a2d-f40e-1e5a76d25567"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['bat_resnext26ts',\n",
       " 'cspresnet50',\n",
       " 'cspresnet50d',\n",
       " 'cspresnet50w',\n",
       " 'cspresnext50',\n",
       " 'cspresnext50_iabn',\n",
       " 'dla60_res2net',\n",
       " 'dla60_res2next',\n",
       " 'eca_resnet33ts',\n",
       " 'eca_resnext26ts',\n",
       " 'ecaresnet26t',\n",
       " 'ecaresnet50d',\n",
       " 'ecaresnet50d_pruned',\n",
       " 'ecaresnet50t',\n",
       " 'ecaresnet101d',\n",
       " 'ecaresnet101d_pruned',\n",
       " 'ecaresnet200d',\n",
       " 'ecaresnet269d',\n",
       " 'ecaresnetlight',\n",
       " 'ecaresnext26t_32x4d',\n",
       " 'ecaresnext50t_32x4d',\n",
       " 'ens_adv_inception_resnet_v2',\n",
       " 'gcresnet33ts',\n",
       " 'gcresnet50t',\n",
       " 'gcresnext26ts',\n",
       " 'gcresnext50ts',\n",
       " 'gluon_resnet18_v1b',\n",
       " 'gluon_resnet34_v1b',\n",
       " 'gluon_resnet50_v1b',\n",
       " 'gluon_resnet50_v1c',\n",
       " 'gluon_resnet50_v1d',\n",
       " 'gluon_resnet50_v1s',\n",
       " 'gluon_resnet101_v1b',\n",
       " 'gluon_resnet101_v1c',\n",
       " 'gluon_resnet101_v1d',\n",
       " 'gluon_resnet101_v1s',\n",
       " 'gluon_resnet152_v1b',\n",
       " 'gluon_resnet152_v1c',\n",
       " 'gluon_resnet152_v1d',\n",
       " 'gluon_resnet152_v1s',\n",
       " 'gluon_resnext50_32x4d',\n",
       " 'gluon_resnext101_32x4d',\n",
       " 'gluon_resnext101_64x4d',\n",
       " 'gluon_seresnext50_32x4d',\n",
       " 'gluon_seresnext101_32x4d',\n",
       " 'gluon_seresnext101_64x4d',\n",
       " 'ig_resnext101_32x8d',\n",
       " 'ig_resnext101_32x16d',\n",
       " 'ig_resnext101_32x32d',\n",
       " 'ig_resnext101_32x48d',\n",
       " 'inception_resnet_v2',\n",
       " 'lambda_resnet26rpt_256',\n",
       " 'lambda_resnet26t',\n",
       " 'lambda_resnet50ts',\n",
       " 'legacy_seresnet18',\n",
       " 'legacy_seresnet34',\n",
       " 'legacy_seresnet50',\n",
       " 'legacy_seresnet101',\n",
       " 'legacy_seresnet152',\n",
       " 'legacy_seresnext26_32x4d',\n",
       " 'legacy_seresnext50_32x4d',\n",
       " 'legacy_seresnext101_32x4d',\n",
       " 'nf_ecaresnet26',\n",
       " 'nf_ecaresnet50',\n",
       " 'nf_ecaresnet101',\n",
       " 'nf_resnet26',\n",
       " 'nf_resnet50',\n",
       " 'nf_resnet101',\n",
       " 'nf_seresnet26',\n",
       " 'nf_seresnet50',\n",
       " 'nf_seresnet101',\n",
       " 'res2net50_14w_8s',\n",
       " 'res2net50_26w_4s',\n",
       " 'res2net50_26w_6s',\n",
       " 'res2net50_26w_8s',\n",
       " 'res2net50_48w_2s',\n",
       " 'res2net101_26w_4s',\n",
       " 'res2next50',\n",
       " 'resmlp_12_224',\n",
       " 'resmlp_12_224_dino',\n",
       " 'resmlp_12_distilled_224',\n",
       " 'resmlp_24_224',\n",
       " 'resmlp_24_224_dino',\n",
       " 'resmlp_24_distilled_224',\n",
       " 'resmlp_36_224',\n",
       " 'resmlp_36_distilled_224',\n",
       " 'resmlp_big_24_224',\n",
       " 'resmlp_big_24_224_in22ft1k',\n",
       " 'resmlp_big_24_distilled_224',\n",
       " 'resnest14d',\n",
       " 'resnest26d',\n",
       " 'resnest50d',\n",
       " 'resnest50d_1s4x24d',\n",
       " 'resnest50d_4s2x40d',\n",
       " 'resnest101e',\n",
       " 'resnest200e',\n",
       " 'resnest269e',\n",
       " 'resnet18',\n",
       " 'resnet18d',\n",
       " 'resnet26',\n",
       " 'resnet26d',\n",
       " 'resnet26t',\n",
       " 'resnet32ts',\n",
       " 'resnet33ts',\n",
       " 'resnet34',\n",
       " 'resnet34d',\n",
       " 'resnet50',\n",
       " 'resnet50_gn',\n",
       " 'resnet50d',\n",
       " 'resnet50t',\n",
       " 'resnet51q',\n",
       " 'resnet61q',\n",
       " 'resnet101',\n",
       " 'resnet101d',\n",
       " 'resnet152',\n",
       " 'resnet152d',\n",
       " 'resnet200',\n",
       " 'resnet200d',\n",
       " 'resnetblur18',\n",
       " 'resnetblur50',\n",
       " 'resnetrs50',\n",
       " 'resnetrs101',\n",
       " 'resnetrs152',\n",
       " 'resnetrs200',\n",
       " 'resnetrs270',\n",
       " 'resnetrs350',\n",
       " 'resnetrs420',\n",
       " 'resnetv2_50',\n",
       " 'resnetv2_50d',\n",
       " 'resnetv2_50d_evob',\n",
       " 'resnetv2_50d_evos',\n",
       " 'resnetv2_50d_gn',\n",
       " 'resnetv2_50t',\n",
       " 'resnetv2_50x1_bit_distilled',\n",
       " 'resnetv2_50x1_bitm',\n",
       " 'resnetv2_50x1_bitm_in21k',\n",
       " 'resnetv2_50x3_bitm',\n",
       " 'resnetv2_50x3_bitm_in21k',\n",
       " 'resnetv2_101',\n",
       " 'resnetv2_101d',\n",
       " 'resnetv2_101x1_bitm',\n",
       " 'resnetv2_101x1_bitm_in21k',\n",
       " 'resnetv2_101x3_bitm',\n",
       " 'resnetv2_101x3_bitm_in21k',\n",
       " 'resnetv2_152',\n",
       " 'resnetv2_152d',\n",
       " 'resnetv2_152x2_bit_teacher',\n",
       " 'resnetv2_152x2_bit_teacher_384',\n",
       " 'resnetv2_152x2_bitm',\n",
       " 'resnetv2_152x2_bitm_in21k',\n",
       " 'resnetv2_152x4_bitm',\n",
       " 'resnetv2_152x4_bitm_in21k',\n",
       " 'resnext26ts',\n",
       " 'resnext50_32x4d',\n",
       " 'resnext50d_32x4d',\n",
       " 'resnext101_32x4d',\n",
       " 'resnext101_32x8d',\n",
       " 'resnext101_64x4d',\n",
       " 'seresnet18',\n",
       " 'seresnet33ts',\n",
       " 'seresnet34',\n",
       " 'seresnet50',\n",
       " 'seresnet50t',\n",
       " 'seresnet101',\n",
       " 'seresnet152',\n",
       " 'seresnet152d',\n",
       " 'seresnet200d',\n",
       " 'seresnet269d',\n",
       " 'seresnext26d_32x4d',\n",
       " 'seresnext26t_32x4d',\n",
       " 'seresnext26tn_32x4d',\n",
       " 'seresnext26ts',\n",
       " 'seresnext50_32x4d',\n",
       " 'seresnext101_32x4d',\n",
       " 'seresnext101_32x8d',\n",
       " 'skresnet18',\n",
       " 'skresnet34',\n",
       " 'skresnet50',\n",
       " 'skresnet50d',\n",
       " 'skresnext50_32x4d',\n",
       " 'ssl_resnet18',\n",
       " 'ssl_resnet50',\n",
       " 'ssl_resnext50_32x4d',\n",
       " 'ssl_resnext101_32x4d',\n",
       " 'ssl_resnext101_32x8d',\n",
       " 'ssl_resnext101_32x16d',\n",
       " 'swsl_resnet18',\n",
       " 'swsl_resnet50',\n",
       " 'swsl_resnext50_32x4d',\n",
       " 'swsl_resnext101_32x4d',\n",
       " 'swsl_resnext101_32x8d',\n",
       " 'swsl_resnext101_32x16d',\n",
       " 'tresnet_l',\n",
       " 'tresnet_l_448',\n",
       " 'tresnet_m',\n",
       " 'tresnet_m_448',\n",
       " 'tresnet_m_miil_in21k',\n",
       " 'tresnet_xl',\n",
       " 'tresnet_xl_448',\n",
       " 'tv_resnet34',\n",
       " 'tv_resnet50',\n",
       " 'tv_resnet101',\n",
       " 'tv_resnet152',\n",
       " 'tv_resnext50_32x4d',\n",
       " 'vit_base_resnet26d_224',\n",
       " 'vit_base_resnet50_224_in21k',\n",
       " 'vit_base_resnet50_384',\n",
       " 'vit_base_resnet50d_224',\n",
       " 'vit_small_resnet26d_224',\n",
       " 'vit_small_resnet50d_s16_224',\n",
       " 'wide_resnet50_2',\n",
       " 'wide_resnet101_2']"
      ]
     },
     "execution_count": 161,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## efficientnet 목록\n",
    "timm.list_models('*res*')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 359,
   "metadata": {
    "id": "I3xxQtIK88jr"
   },
   "outputs": [],
   "source": [
    " ## 신경망 fine-tunning\n",
    "class EffNet(nn.Module):\n",
    "  def __init__(self,num_classes):\n",
    "    super(EffNet, self).__init__()\n",
    "    \n",
    "    if os.path.isfile('models/best_model.pt'):\n",
    "        print('BEST MODEL LOAD')\n",
    "        self.model = torch.load('models/best_model.pt')\n",
    "        \n",
    "    else:\n",
    "        print('PRE-TRAINED MODEL LOAD')\n",
    "        self.model = timm.create_model('resnetv2_101', pretrained = True)\n",
    "\n",
    "        # num_features = self.model.fc.in_features\n",
    "        self.model.classifier = nn.Sequential(\n",
    "            nn.Linear(in_features = 2048, out_features = 640),\n",
    "            nn.ReLU(),\n",
    "            nn.Dropout(p = DROPOUT_RATE),\n",
    "            nn.Linear(in_features = 640, out_features = 256),\n",
    "            nn.ReLU(),\n",
    "            nn.Dropout(p = DROPOUT_RATE),\n",
    "            nn.Linear(in_features = 256, out_features = 64),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(in_features = 64, out_features = num_classes)\n",
    "        )\n",
    "\n",
    "    \n",
    "  def forward(self, x):\n",
    "    output = self.model(x)\n",
    "    return output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {
    "id": "ANl75jv7VJGV"
   },
   "outputs": [],
   "source": [
    "## 손실함수 정의\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "\n",
    "## 최적화 함수 가져오는 함수\n",
    "def get_optim(model):\n",
    "  optim = AdamW(model.parameters(), lr = LR, weight_decay = 5e-6)\n",
    "\n",
    "  ## 모든 파라미터들의 grad값을 0으로 초기화\n",
    "  optim.zero_grad()\n",
    "  return optim\n",
    "\n",
    "## 러닝레이트 스케줄러함수\n",
    "def get_scheduler(optim):\n",
    "  scheduler = CosineAnnealingWarmupRestarts(optim, first_cycle_steps = 10, cycle_mult = 1, max_lr = 0.1, min_lr = 0.0001, warmup_steps = 5, gamma = 0.5)\n",
    "  return scheduler\n",
    "\n",
    "## 학습 함수\n",
    "def train(model, train_loader, optim):\n",
    "  model.train()\n",
    "  corrects, total_loss = 0, 0\n",
    " \n",
    "  for (images, labels) in train_loader:\n",
    "    images, labels = images.to(DEVICE), labels.to(DEVICE)\n",
    "\n",
    "    outputs = model(images)\n",
    "    loss = criterion(outputs, labels)\n",
    "    loss.backward()\n",
    "\n",
    "    optim.step()\n",
    "    optim.zero_grad()\n",
    "\n",
    "    _, preds = torch.max(outputs, 1)\n",
    "    corrects += torch.sum(preds == labels.data)\n",
    "    total_loss += loss.item()\n",
    "\n",
    "  avg_loss = total_loss / len(train_loader)\n",
    "  acc = corrects / (len(train_loader)*BATCH_SIZE)\n",
    "\n",
    "  print(f'Train Accuracy : {acc:.3f} | Train Loss : {avg_loss:.3f}')\n",
    "  return acc\n",
    "\n",
    "## 검증 함수\n",
    "def validation(model, valid_loader):\n",
    "  model.eval()\n",
    "  corrects, total_loss = 0, 0\n",
    "\n",
    "  for (images, labels) in valid_loader:\n",
    "    images, labels = images.to(DEVICE), labels.to(DEVICE)\n",
    "\n",
    "    outputs = model(images)\n",
    "    loss = criterion(outputs, labels)\n",
    "\n",
    "    _, preds = torch.max(outputs, 1)\n",
    "    corrects += torch.sum(preds == labels.data)\n",
    "    total_loss += loss.item()\n",
    "\n",
    "  avg_loss = total_loss / len(valid_loader)\n",
    "  acc = corrects / (len(valid_loader)*BATCH_SIZE)\n",
    "  print(f'Valid Accuracy : {acc:.3f} | Valid Loss : {avg_loss:.3f}\\n')\n",
    "\n",
    "  return acc, outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 361,
   "metadata": {
    "id": "SHI7-lQboOBv"
   },
   "outputs": [],
   "source": [
    "## 훈련 사이클 함수\n",
    "def run(model, train_loader, valid_loader):\n",
    "  optim     = get_optim(model)\n",
    "  # scheduler = get_scheduler(optim)\n",
    "\n",
    "  early_stopping_counter = 0\n",
    "  best_acc               = -1\n",
    "\n",
    "  for epoch in range(EPOCHS):\n",
    "    print(f'[{epoch + 1} / {EPOCHS}]')\n",
    "\n",
    "    train_acc          = train(model, train_loader, optim)\n",
    "    valid_acc, outputs = validation(model, valid_loader)\n",
    "\n",
    "    if valid_acc > best_acc:\n",
    "        os.makedirs('models/', exist_ok = True)\n",
    "        \n",
    "        best_acc               = valid_acc\n",
    "        early_stopping_counter = 0\n",
    "        \n",
    "        torch.save(model, 'models/best_model.pt')\n",
    "        print(f'save model with accuracy : {best_acc}')\n",
    "\n",
    "    else:\n",
    "        early_stopping_counter += 1\n",
    "        if early_stopping_counter >= PATIENCE:\n",
    "            print(f'Training Stopped by Early stopping counter : {early_stopping_counter}')\n",
    "            break\n",
    "\n",
    "    optim.step()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 362,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dc4RtywI8tOG",
    "outputId": "bd7c2c80-9746-4c74-c5a2-474361e5b23f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PRE-TRAINED MODEL LOAD\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "EffNet(\n",
       "  (model): ResNetV2(\n",
       "    (stem): Sequential(\n",
       "      (conv): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
       "      (pool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "    )\n",
       "    (stages): Sequential(\n",
       "      (0): ResNetStage(\n",
       "        (blocks): Sequential(\n",
       "          (0): PreActBottleneck(\n",
       "            (downsample): DownsampleConv(\n",
       "              (conv): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "              (norm): Identity()\n",
       "            )\n",
       "            (norm1): BatchNormAct2d(\n",
       "              64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (1): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (2): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "        )\n",
       "      )\n",
       "      (1): ResNetStage(\n",
       "        (blocks): Sequential(\n",
       "          (0): PreActBottleneck(\n",
       "            (downsample): DownsampleConv(\n",
       "              (conv): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "              (norm): Identity()\n",
       "            )\n",
       "            (norm1): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (1): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (2): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (3): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "        )\n",
       "      )\n",
       "      (2): ResNetStage(\n",
       "        (blocks): Sequential(\n",
       "          (0): PreActBottleneck(\n",
       "            (downsample): DownsampleConv(\n",
       "              (conv): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "              (norm): Identity()\n",
       "            )\n",
       "            (norm1): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (1): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (2): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (3): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (4): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (5): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (6): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (7): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (8): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (9): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (10): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (11): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (12): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (13): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (14): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (15): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (16): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (17): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (18): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (19): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (20): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (21): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (22): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "        )\n",
       "      )\n",
       "      (3): ResNetStage(\n",
       "        (blocks): Sequential(\n",
       "          (0): PreActBottleneck(\n",
       "            (downsample): DownsampleConv(\n",
       "              (conv): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "              (norm): Identity()\n",
       "            )\n",
       "            (norm1): BatchNormAct2d(\n",
       "              1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (1): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "          (2): PreActBottleneck(\n",
       "            (norm1): BatchNormAct2d(\n",
       "              2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (norm2): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "            (norm3): BatchNormAct2d(\n",
       "              512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "              (act): ReLU(inplace=True)\n",
       "            )\n",
       "            (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (drop_path): Identity()\n",
       "          )\n",
       "        )\n",
       "      )\n",
       "    )\n",
       "    (norm): BatchNormAct2d(\n",
       "      2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True\n",
       "      (act): ReLU(inplace=True)\n",
       "    )\n",
       "    (head): ClassifierHead(\n",
       "      (global_pool): SelectAdaptivePool2d (pool_type=avg, flatten=Identity())\n",
       "      (fc): Conv2d(2048, 1000, kernel_size=(1, 1), stride=(1, 1))\n",
       "      (flatten): Flatten(start_dim=1, end_dim=-1)\n",
       "    )\n",
       "    (classifier): Sequential(\n",
       "      (0): Linear(in_features=2048, out_features=640, bias=True)\n",
       "      (1): ReLU()\n",
       "      (2): Dropout(p=0.5, inplace=False)\n",
       "      (3): Linear(in_features=640, out_features=256, bias=True)\n",
       "      (4): ReLU()\n",
       "      (5): Dropout(p=0.5, inplace=False)\n",
       "      (6): Linear(in_features=256, out_features=64, bias=True)\n",
       "      (7): ReLU()\n",
       "      (8): Linear(in_features=64, out_features=5, bias=True)\n",
       "    )\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 362,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## 캐시 메모리 비우기 및 가비지 컬렉터 실행\n",
    "## 모델 여러번 학습시키고 나면 메모리에 남는 찌꺼기들 제거해주는 코드\n",
    "torch.cuda.empty_cache()\n",
    "gc.collect()\n",
    "\n",
    "\n",
    "model = EffNet(num_classes = NUM_CLASSES)\n",
    "model.to(DEVICE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 353,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 401
    },
    "id": "_dQBlpMfum0P",
    "outputId": "6539ea7a-3e81-4b3b-efb6-2749804ff5cd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 / 100]\n",
      "Train Accuracy : 0.386 | Train Loss : 1.626\n",
      "Valid Accuracy : 0.390 | Valid Loss : 1.314\n",
      "\n",
      "save model with accuracy : 0.3898809850215912\n",
      "[2 / 100]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Input \u001b[0;32mIn [353]\u001b[0m, in \u001b[0;36m<cell line: 1>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtrain_loader\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mvalid_loader\u001b[49m\u001b[43m)\u001b[49m\n",
      "Input \u001b[0;32mIn [351]\u001b[0m, in \u001b[0;36mrun\u001b[0;34m(model, train_loader, valid_loader)\u001b[0m\n\u001b[1;32m      9\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m epoch \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(EPOCHS):\n\u001b[1;32m     10\u001b[0m   \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m[\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mepoch \u001b[38;5;241m+\u001b[39m \u001b[38;5;241m1\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m / \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mEPOCHS\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m]\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m---> 12\u001b[0m   train_acc          \u001b[38;5;241m=\u001b[39m \u001b[43mtrain\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtrain_loader\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43moptim\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     13\u001b[0m   valid_acc, outputs \u001b[38;5;241m=\u001b[39m validation(model, valid_loader)\n\u001b[1;32m     15\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m valid_acc \u001b[38;5;241m>\u001b[39m best_acc:\n",
      "Input \u001b[0;32mIn [350]\u001b[0m, in \u001b[0;36mtrain\u001b[0;34m(model, train_loader, optim)\u001b[0m\n\u001b[1;32m     25\u001b[0m outputs \u001b[38;5;241m=\u001b[39m model(images)\n\u001b[1;32m     26\u001b[0m loss \u001b[38;5;241m=\u001b[39m criterion(outputs, labels)\n\u001b[0;32m---> 27\u001b[0m \u001b[43mloss\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     29\u001b[0m optim\u001b[38;5;241m.\u001b[39mstep()\n\u001b[1;32m     30\u001b[0m optim\u001b[38;5;241m.\u001b[39mzero_grad()\n",
      "File \u001b[0;32m/opt/conda/envs/bio/lib/python3.9/site-packages/torch/_tensor.py:307\u001b[0m, in \u001b[0;36mTensor.backward\u001b[0;34m(self, gradient, retain_graph, create_graph, inputs)\u001b[0m\n\u001b[1;32m    298\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m has_torch_function_unary(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m    299\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m handle_torch_function(\n\u001b[1;32m    300\u001b[0m         Tensor\u001b[38;5;241m.\u001b[39mbackward,\n\u001b[1;32m    301\u001b[0m         (\u001b[38;5;28mself\u001b[39m,),\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    305\u001b[0m         create_graph\u001b[38;5;241m=\u001b[39mcreate_graph,\n\u001b[1;32m    306\u001b[0m         inputs\u001b[38;5;241m=\u001b[39minputs)\n\u001b[0;32m--> 307\u001b[0m \u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mautograd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgradient\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minputs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/envs/bio/lib/python3.9/site-packages/torch/autograd/__init__.py:154\u001b[0m, in \u001b[0;36mbackward\u001b[0;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[0m\n\u001b[1;32m    151\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m retain_graph \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    152\u001b[0m     retain_graph \u001b[38;5;241m=\u001b[39m create_graph\n\u001b[0;32m--> 154\u001b[0m \u001b[43mVariable\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_execution_engine\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun_backward\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    155\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtensors\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgrad_tensors_\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    156\u001b[0m \u001b[43m    \u001b[49m\u001b[43mallow_unreachable\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maccumulate_grad\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "run(model, train_loader, valid_loader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "eTK7wMmHuzV2"
   },
   "outputs": [],
   "source": [
    " from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "id": "hzltYuIXvES_"
   },
   "outputs": [],
   "source": [
    "test_dataset = BeafData(test_images, test_labels, dtype = 'test')\n",
    "test_loader  = DataLoader(test_dataset, batch_size = 1, shuffle = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "I_pY6QY_FGPa",
    "outputId": "05d08f27-e48d-4cf7-c940-da82183dc7de"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy : 0.712\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         1++       0.71      0.61      0.66       440\n",
      "          1+       0.63      0.60      0.61       427\n",
      "           1       0.58      0.75      0.66       427\n",
      "           2       0.87      0.72      0.79       418\n",
      "           3       0.87      0.96      0.91       288\n",
      "\n",
      "    accuracy                           0.71      2000\n",
      "   macro avg       0.73      0.73      0.73      2000\n",
      "weighted avg       0.72      0.71      0.71      2000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "## 성능 지표 계산함수\n",
    "def evaluation(model, test_loader):\n",
    "  model.eval()\n",
    "  corrects = 0\n",
    "  preds, gt = [], []\n",
    "\n",
    "  for images, labels in test_loader:\n",
    "    images, labels = images.to(DEVICE), labels.to(DEVICE)\n",
    "\n",
    "    outputs = model(images)\n",
    "\n",
    "    _, pred = torch.max(outputs, 1)\n",
    "    corrects += torch.sum(pred == labels.data)\n",
    "\n",
    "    preds.append(pred.item())\n",
    "    gt.append(labels.data.item())\n",
    "\n",
    "  acc = corrects / len(test_loader)\n",
    "  print(f'Test Accuracy : {acc:.3f}\\n')\n",
    "\n",
    "  return classification_report(gt, preds, target_names = label2idx.keys())\n",
    "\n",
    "report = evaluation(model, test_loader)\n",
    "print(report)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "cETZJ9g3FMqI"
   },
   "outputs": [],
   "source": [
    "-"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Beaf_classification.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "bio",
   "language": "python",
   "name": "bio"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}