timm_classif_cut_davit_tiny.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ef58f141-4122-4d78-b65e-9d7d06817682",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import time\n",
    "import torch\n",
    "import timm\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from torchvision import datasets, transforms\n",
    "from torch.utils.data import DataLoader, random_split\n",
    "from sklearn.model_selection import train_test_split\n",
    "from datetime import datetime\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8f1ee3b9-67a6-442e-806a-a9057064da54",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set device\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "device\n",
    "\n",
    "# Custom dataset folder\n",
    "#dataset_path = 'F:/vit_custom_img_classif/key_desc_org'  # Update this path\n",
    "dataset_path = 'dataset_cut'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b7346cb7-43d4-425c-8d87-7a13585dcdf9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preset values\n",
    "dropout_rate = 0.00\n",
    "batch_size = 1\n",
    "learning_rate = 1e-4\n",
    "num_epochs = 40\n",
    "image_size = 514"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "80cc39cd-5f4c-4fb0-acef-59659ce4867e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data augmentation and normalization for training\n",
    "transform = transforms.Compose([\n",
    "    transforms.Resize((image_size, image_size)),\n",
    "    #transforms.RandomHorizontalFlip(),\n",
    "    #transforms.RandomRotation(degrees=(0, 360)),\n",
    "    #transforms.ColorJitter(brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5),\n",
    "    #transforms.RandomResizedCrop(size=image_size),\n",
    "    #transforms.RandomPerspective(distortion_scale=0.5, probability=0.5),\n",
    "    #transforms.RandomVerticalFlip(),\n",
    "    #transforms.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n",
    "    transforms.RandomErasing(p=0.0),\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "77f1d768-2bc6-4af1-9937-fd2332939ebb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load custom dataset\n",
    "dataset = datasets.ImageFolder(dataset_path, transform=transform)\n",
    "class_names = dataset.classes\n",
    "num_classes = len(class_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "03e6ddb5-5cd5-4707-968d-7bd1689f1018",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split dataset into training and validation sets\n",
    "train_size = int(0.7 * len(dataset))\n",
    "val_size = len(dataset) - train_size\n",
    "train_dataset, val_dataset = random_split(dataset, [train_size, val_size])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ea711ecf-5d86-4cb9-abaf-ec7283253ddb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create data loaders\n",
    "train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)\n",
    "val_loader = DataLoader(val_dataset, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3fda9f83-306e-45ad-9d7f-89ed8b7b55ad",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['bat_resnext26ts',\n",
       " 'beit_base_patch16_224',\n",
       " 'beit_base_patch16_384',\n",
       " 'beit_large_patch16_224',\n",
       " 'beit_large_patch16_384',\n",
       " 'beit_large_patch16_512',\n",
       " 'beitv2_base_patch16_224',\n",
       " 'beitv2_large_patch16_224',\n",
       " 'botnet26t_256',\n",
       " 'botnet50ts_256',\n",
       " 'caformer_b36',\n",
       " 'caformer_m36',\n",
       " 'caformer_s18',\n",
       " 'caformer_s36',\n",
       " 'cait_m36_384',\n",
       " 'cait_m48_448',\n",
       " 'cait_s24_224',\n",
       " 'cait_s24_384',\n",
       " 'cait_s36_384',\n",
       " 'cait_xs24_384',\n",
       " 'cait_xxs24_224',\n",
       " 'cait_xxs24_384',\n",
       " 'cait_xxs36_224',\n",
       " 'cait_xxs36_384',\n",
       " 'coat_lite_medium',\n",
       " 'coat_lite_medium_384',\n",
       " 'coat_lite_mini',\n",
       " 'coat_lite_small',\n",
       " 'coat_lite_tiny',\n",
       " 'coat_mini',\n",
       " 'coat_small',\n",
       " 'coat_tiny',\n",
       " 'coatnet_0_224',\n",
       " 'coatnet_0_rw_224',\n",
       " 'coatnet_1_224',\n",
       " 'coatnet_1_rw_224',\n",
       " 'coatnet_2_224',\n",
       " 'coatnet_2_rw_224',\n",
       " 'coatnet_3_224',\n",
       " 'coatnet_3_rw_224',\n",
       " 'coatnet_4_224',\n",
       " 'coatnet_5_224',\n",
       " 'coatnet_bn_0_rw_224',\n",
       " 'coatnet_nano_cc_224',\n",
       " 'coatnet_nano_rw_224',\n",
       " 'coatnet_pico_rw_224',\n",
       " 'coatnet_rmlp_0_rw_224',\n",
       " 'coatnet_rmlp_1_rw2_224',\n",
       " 'coatnet_rmlp_1_rw_224',\n",
       " 'coatnet_rmlp_2_rw_224',\n",
       " 'coatnet_rmlp_2_rw_384',\n",
       " 'coatnet_rmlp_3_rw_224',\n",
       " 'coatnet_rmlp_nano_rw_224',\n",
       " 'coatnext_nano_rw_224',\n",
       " 'convformer_b36',\n",
       " 'convformer_m36',\n",
       " 'convformer_s18',\n",
       " 'convformer_s36',\n",
       " 'convit_base',\n",
       " 'convit_small',\n",
       " 'convit_tiny',\n",
       " 'convmixer_768_32',\n",
       " 'convmixer_1024_20_ks9_p14',\n",
       " 'convmixer_1536_20',\n",
       " 'convnext_atto',\n",
       " 'convnext_atto_ols',\n",
       " 'convnext_base',\n",
       " 'convnext_femto',\n",
       " 'convnext_femto_ols',\n",
       " 'convnext_large',\n",
       " 'convnext_large_mlp',\n",
       " 'convnext_nano',\n",
       " 'convnext_nano_ols',\n",
       " 'convnext_pico',\n",
       " 'convnext_pico_ols',\n",
       " 'convnext_small',\n",
       " 'convnext_tiny',\n",
       " 'convnext_tiny_hnf',\n",
       " 'convnext_xlarge',\n",
       " 'convnext_xxlarge',\n",
       " 'convnextv2_atto',\n",
       " 'convnextv2_base',\n",
       " 'convnextv2_femto',\n",
       " 'convnextv2_huge',\n",
       " 'convnextv2_large',\n",
       " 'convnextv2_nano',\n",
       " 'convnextv2_pico',\n",
       " 'convnextv2_small',\n",
       " 'convnextv2_tiny',\n",
       " 'crossvit_9_240',\n",
       " 'crossvit_9_dagger_240',\n",
       " 'crossvit_15_240',\n",
       " 'crossvit_15_dagger_240',\n",
       " 'crossvit_15_dagger_408',\n",
       " 'crossvit_18_240',\n",
       " 'crossvit_18_dagger_240',\n",
       " 'crossvit_18_dagger_408',\n",
       " 'crossvit_base_240',\n",
       " 'crossvit_small_240',\n",
       " 'crossvit_tiny_240',\n",
       " 'cs3darknet_focus_l',\n",
       " 'cs3darknet_focus_m',\n",
       " 'cs3darknet_focus_s',\n",
       " 'cs3darknet_focus_x',\n",
       " 'cs3darknet_l',\n",
       " 'cs3darknet_m',\n",
       " 'cs3darknet_s',\n",
       " 'cs3darknet_x',\n",
       " 'cs3edgenet_x',\n",
       " 'cs3se_edgenet_x',\n",
       " 'cs3sedarknet_l',\n",
       " 'cs3sedarknet_x',\n",
       " 'cs3sedarknet_xdw',\n",
       " 'cspdarknet53',\n",
       " 'cspresnet50',\n",
       " 'cspresnet50d',\n",
       " 'cspresnet50w',\n",
       " 'cspresnext50',\n",
       " 'darknet17',\n",
       " 'darknet21',\n",
       " 'darknet53',\n",
       " 'darknetaa53',\n",
       " 'davit_base',\n",
       " 'davit_giant',\n",
       " 'davit_huge',\n",
       " 'davit_large',\n",
       " 'davit_small',\n",
       " 'davit_tiny',\n",
       " 'deit3_base_patch16_224',\n",
       " 'deit3_base_patch16_384',\n",
       " 'deit3_huge_patch14_224',\n",
       " 'deit3_large_patch16_224',\n",
       " 'deit3_large_patch16_384',\n",
       " 'deit3_medium_patch16_224',\n",
       " 'deit3_small_patch16_224',\n",
       " 'deit3_small_patch16_384',\n",
       " 'deit_base_distilled_patch16_224',\n",
       " 'deit_base_distilled_patch16_384',\n",
       " 'deit_base_patch16_224',\n",
       " 'deit_base_patch16_384',\n",
       " 'deit_small_distilled_patch16_224',\n",
       " 'deit_small_patch16_224',\n",
       " 'deit_tiny_distilled_patch16_224',\n",
       " 'deit_tiny_patch16_224',\n",
       " 'densenet121',\n",
       " 'densenet161',\n",
       " 'densenet169',\n",
       " 'densenet201',\n",
       " 'densenet264d',\n",
       " 'densenetblur121d',\n",
       " 'dla34',\n",
       " 'dla46_c',\n",
       " 'dla46x_c',\n",
       " 'dla60',\n",
       " 'dla60_res2net',\n",
       " 'dla60_res2next',\n",
       " 'dla60x',\n",
       " 'dla60x_c',\n",
       " 'dla102',\n",
       " 'dla102x',\n",
       " 'dla102x2',\n",
       " 'dla169',\n",
       " 'dm_nfnet_f0',\n",
       " 'dm_nfnet_f1',\n",
       " 'dm_nfnet_f2',\n",
       " 'dm_nfnet_f3',\n",
       " 'dm_nfnet_f4',\n",
       " 'dm_nfnet_f5',\n",
       " 'dm_nfnet_f6',\n",
       " 'dpn48b',\n",
       " 'dpn68',\n",
       " 'dpn68b',\n",
       " 'dpn92',\n",
       " 'dpn98',\n",
       " 'dpn107',\n",
       " 'dpn131',\n",
       " 'eca_botnext26ts_256',\n",
       " 'eca_halonext26ts',\n",
       " 'eca_nfnet_l0',\n",
       " 'eca_nfnet_l1',\n",
       " 'eca_nfnet_l2',\n",
       " 'eca_nfnet_l3',\n",
       " 'eca_resnet33ts',\n",
       " 'eca_resnext26ts',\n",
       " 'eca_vovnet39b',\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",
       " 'edgenext_base',\n",
       " 'edgenext_small',\n",
       " 'edgenext_small_rw',\n",
       " 'edgenext_x_small',\n",
       " 'edgenext_xx_small',\n",
       " 'efficientformer_l1',\n",
       " 'efficientformer_l3',\n",
       " 'efficientformer_l7',\n",
       " 'efficientformerv2_l',\n",
       " 'efficientformerv2_s0',\n",
       " 'efficientformerv2_s1',\n",
       " 'efficientformerv2_s2',\n",
       " 'efficientnet_b0',\n",
       " 'efficientnet_b0_g8_gn',\n",
       " 'efficientnet_b0_g16_evos',\n",
       " 'efficientnet_b0_gn',\n",
       " 'efficientnet_b1',\n",
       " 'efficientnet_b1_pruned',\n",
       " 'efficientnet_b2',\n",
       " 'efficientnet_b2_pruned',\n",
       " 'efficientnet_b3',\n",
       " 'efficientnet_b3_g8_gn',\n",
       " 'efficientnet_b3_gn',\n",
       " 'efficientnet_b3_pruned',\n",
       " 'efficientnet_b4',\n",
       " 'efficientnet_b5',\n",
       " 'efficientnet_b6',\n",
       " 'efficientnet_b7',\n",
       " 'efficientnet_b8',\n",
       " 'efficientnet_cc_b0_4e',\n",
       " 'efficientnet_cc_b0_8e',\n",
       " 'efficientnet_cc_b1_8e',\n",
       " 'efficientnet_el',\n",
       " 'efficientnet_el_pruned',\n",
       " 'efficientnet_em',\n",
       " 'efficientnet_es',\n",
       " 'efficientnet_es_pruned',\n",
       " 'efficientnet_l2',\n",
       " 'efficientnet_lite0',\n",
       " 'efficientnet_lite1',\n",
       " 'efficientnet_lite2',\n",
       " 'efficientnet_lite3',\n",
       " 'efficientnet_lite4',\n",
       " 'efficientnetv2_l',\n",
       " 'efficientnetv2_m',\n",
       " 'efficientnetv2_rw_m',\n",
       " 'efficientnetv2_rw_s',\n",
       " 'efficientnetv2_rw_t',\n",
       " 'efficientnetv2_s',\n",
       " 'efficientnetv2_xl',\n",
       " 'efficientvit_b0',\n",
       " 'efficientvit_b1',\n",
       " 'efficientvit_b2',\n",
       " 'efficientvit_b3',\n",
       " 'efficientvit_l1',\n",
       " 'efficientvit_l2',\n",
       " 'efficientvit_l3',\n",
       " 'efficientvit_m0',\n",
       " 'efficientvit_m1',\n",
       " 'efficientvit_m2',\n",
       " 'efficientvit_m3',\n",
       " 'efficientvit_m4',\n",
       " 'efficientvit_m5',\n",
       " 'ese_vovnet19b_dw',\n",
       " 'ese_vovnet19b_slim',\n",
       " 'ese_vovnet19b_slim_dw',\n",
       " 'ese_vovnet39b',\n",
       " 'ese_vovnet39b_evos',\n",
       " 'ese_vovnet57b',\n",
       " 'ese_vovnet99b',\n",
       " 'eva02_base_patch14_224',\n",
       " 'eva02_base_patch14_448',\n",
       " 'eva02_base_patch16_clip_224',\n",
       " 'eva02_enormous_patch14_clip_224',\n",
       " 'eva02_large_patch14_224',\n",
       " 'eva02_large_patch14_448',\n",
       " 'eva02_large_patch14_clip_224',\n",
       " 'eva02_large_patch14_clip_336',\n",
       " 'eva02_small_patch14_224',\n",
       " 'eva02_small_patch14_336',\n",
       " 'eva02_tiny_patch14_224',\n",
       " 'eva02_tiny_patch14_336',\n",
       " 'eva_giant_patch14_224',\n",
       " 'eva_giant_patch14_336',\n",
       " 'eva_giant_patch14_560',\n",
       " 'eva_giant_patch14_clip_224',\n",
       " 'eva_large_patch14_196',\n",
       " 'eva_large_patch14_336',\n",
       " 'fastvit_ma36',\n",
       " 'fastvit_s12',\n",
       " 'fastvit_sa12',\n",
       " 'fastvit_sa24',\n",
       " 'fastvit_sa36',\n",
       " 'fastvit_t8',\n",
       " 'fastvit_t12',\n",
       " 'fbnetc_100',\n",
       " 'fbnetv3_b',\n",
       " 'fbnetv3_d',\n",
       " 'fbnetv3_g',\n",
       " 'flexivit_base',\n",
       " 'flexivit_large',\n",
       " 'flexivit_small',\n",
       " 'focalnet_base_lrf',\n",
       " 'focalnet_base_srf',\n",
       " 'focalnet_huge_fl3',\n",
       " 'focalnet_huge_fl4',\n",
       " 'focalnet_large_fl3',\n",
       " 'focalnet_large_fl4',\n",
       " 'focalnet_small_lrf',\n",
       " 'focalnet_small_srf',\n",
       " 'focalnet_tiny_lrf',\n",
       " 'focalnet_tiny_srf',\n",
       " 'focalnet_xlarge_fl3',\n",
       " 'focalnet_xlarge_fl4',\n",
       " 'gc_efficientnetv2_rw_t',\n",
       " 'gcresnet33ts',\n",
       " 'gcresnet50t',\n",
       " 'gcresnext26ts',\n",
       " 'gcresnext50ts',\n",
       " 'gcvit_base',\n",
       " 'gcvit_small',\n",
       " 'gcvit_tiny',\n",
       " 'gcvit_xtiny',\n",
       " 'gcvit_xxtiny',\n",
       " 'gernet_l',\n",
       " 'gernet_m',\n",
       " 'gernet_s',\n",
       " 'ghostnet_050',\n",
       " 'ghostnet_100',\n",
       " 'ghostnet_130',\n",
       " 'ghostnetv2_100',\n",
       " 'ghostnetv2_130',\n",
       " 'ghostnetv2_160',\n",
       " 'gmixer_12_224',\n",
       " 'gmixer_24_224',\n",
       " 'gmlp_b16_224',\n",
       " 'gmlp_s16_224',\n",
       " 'gmlp_ti16_224',\n",
       " 'halo2botnet50ts_256',\n",
       " 'halonet26t',\n",
       " 'halonet50ts',\n",
       " 'halonet_h1',\n",
       " 'haloregnetz_b',\n",
       " 'hardcorenas_a',\n",
       " 'hardcorenas_b',\n",
       " 'hardcorenas_c',\n",
       " 'hardcorenas_d',\n",
       " 'hardcorenas_e',\n",
       " 'hardcorenas_f',\n",
       " 'hgnet_base',\n",
       " 'hgnet_small',\n",
       " 'hgnet_tiny',\n",
       " 'hgnetv2_b0',\n",
       " 'hgnetv2_b1',\n",
       " 'hgnetv2_b2',\n",
       " 'hgnetv2_b3',\n",
       " 'hgnetv2_b4',\n",
       " 'hgnetv2_b5',\n",
       " 'hgnetv2_b6',\n",
       " 'hrnet_w18',\n",
       " 'hrnet_w18_small',\n",
       " 'hrnet_w18_small_v2',\n",
       " 'hrnet_w18_ssld',\n",
       " 'hrnet_w30',\n",
       " 'hrnet_w32',\n",
       " 'hrnet_w40',\n",
       " 'hrnet_w44',\n",
       " 'hrnet_w48',\n",
       " 'hrnet_w48_ssld',\n",
       " 'hrnet_w64',\n",
       " 'inception_next_base',\n",
       " 'inception_next_small',\n",
       " 'inception_next_tiny',\n",
       " 'inception_resnet_v2',\n",
       " 'inception_v3',\n",
       " 'inception_v4',\n",
       " 'lambda_resnet26rpt_256',\n",
       " 'lambda_resnet26t',\n",
       " 'lambda_resnet50ts',\n",
       " 'lamhalobotnet50ts_256',\n",
       " 'lcnet_035',\n",
       " 'lcnet_050',\n",
       " 'lcnet_075',\n",
       " 'lcnet_100',\n",
       " 'lcnet_150',\n",
       " 'legacy_senet154',\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",
       " 'legacy_xception',\n",
       " 'levit_128',\n",
       " 'levit_128s',\n",
       " 'levit_192',\n",
       " 'levit_256',\n",
       " 'levit_256d',\n",
       " 'levit_384',\n",
       " 'levit_384_s8',\n",
       " 'levit_512',\n",
       " 'levit_512_s8',\n",
       " 'levit_512d',\n",
       " 'levit_conv_128',\n",
       " 'levit_conv_128s',\n",
       " 'levit_conv_192',\n",
       " 'levit_conv_256',\n",
       " 'levit_conv_256d',\n",
       " 'levit_conv_384',\n",
       " 'levit_conv_384_s8',\n",
       " 'levit_conv_512',\n",
       " 'levit_conv_512_s8',\n",
       " 'levit_conv_512d',\n",
       " 'maxvit_base_tf_224',\n",
       " 'maxvit_base_tf_384',\n",
       " 'maxvit_base_tf_512',\n",
       " 'maxvit_large_tf_224',\n",
       " 'maxvit_large_tf_384',\n",
       " 'maxvit_large_tf_512',\n",
       " 'maxvit_nano_rw_256',\n",
       " 'maxvit_pico_rw_256',\n",
       " 'maxvit_rmlp_base_rw_224',\n",
       " 'maxvit_rmlp_base_rw_384',\n",
       " 'maxvit_rmlp_nano_rw_256',\n",
       " 'maxvit_rmlp_pico_rw_256',\n",
       " 'maxvit_rmlp_small_rw_224',\n",
       " 'maxvit_rmlp_small_rw_256',\n",
       " 'maxvit_rmlp_tiny_rw_256',\n",
       " 'maxvit_small_tf_224',\n",
       " 'maxvit_small_tf_384',\n",
       " 'maxvit_small_tf_512',\n",
       " 'maxvit_tiny_pm_256',\n",
       " 'maxvit_tiny_rw_224',\n",
       " 'maxvit_tiny_rw_256',\n",
       " 'maxvit_tiny_tf_224',\n",
       " 'maxvit_tiny_tf_384',\n",
       " 'maxvit_tiny_tf_512',\n",
       " 'maxvit_xlarge_tf_224',\n",
       " 'maxvit_xlarge_tf_384',\n",
       " 'maxvit_xlarge_tf_512',\n",
       " 'maxxvit_rmlp_nano_rw_256',\n",
       " 'maxxvit_rmlp_small_rw_256',\n",
       " 'maxxvit_rmlp_tiny_rw_256',\n",
       " 'maxxvitv2_nano_rw_256',\n",
       " 'maxxvitv2_rmlp_base_rw_224',\n",
       " 'maxxvitv2_rmlp_base_rw_384',\n",
       " 'maxxvitv2_rmlp_large_rw_224',\n",
       " 'mixer_b16_224',\n",
       " 'mixer_b32_224',\n",
       " 'mixer_l16_224',\n",
       " 'mixer_l32_224',\n",
       " 'mixer_s16_224',\n",
       " 'mixer_s32_224',\n",
       " 'mixnet_l',\n",
       " 'mixnet_m',\n",
       " 'mixnet_s',\n",
       " 'mixnet_xl',\n",
       " 'mixnet_xxl',\n",
       " 'mnasnet_050',\n",
       " 'mnasnet_075',\n",
       " 'mnasnet_100',\n",
       " 'mnasnet_140',\n",
       " 'mnasnet_small',\n",
       " 'mobilenetv2_035',\n",
       " 'mobilenetv2_050',\n",
       " 'mobilenetv2_075',\n",
       " 'mobilenetv2_100',\n",
       " 'mobilenetv2_110d',\n",
       " 'mobilenetv2_120d',\n",
       " 'mobilenetv2_140',\n",
       " 'mobilenetv3_large_075',\n",
       " 'mobilenetv3_large_100',\n",
       " 'mobilenetv3_rw',\n",
       " 'mobilenetv3_small_050',\n",
       " 'mobilenetv3_small_075',\n",
       " 'mobilenetv3_small_100',\n",
       " 'mobileone_s0',\n",
       " 'mobileone_s1',\n",
       " 'mobileone_s2',\n",
       " 'mobileone_s3',\n",
       " 'mobileone_s4',\n",
       " 'mobilevit_s',\n",
       " 'mobilevit_xs',\n",
       " 'mobilevit_xxs',\n",
       " 'mobilevitv2_050',\n",
       " 'mobilevitv2_075',\n",
       " 'mobilevitv2_100',\n",
       " 'mobilevitv2_125',\n",
       " 'mobilevitv2_150',\n",
       " 'mobilevitv2_175',\n",
       " 'mobilevitv2_200',\n",
       " 'mvitv2_base',\n",
       " 'mvitv2_base_cls',\n",
       " 'mvitv2_huge_cls',\n",
       " 'mvitv2_large',\n",
       " 'mvitv2_large_cls',\n",
       " 'mvitv2_small',\n",
       " 'mvitv2_small_cls',\n",
       " 'mvitv2_tiny',\n",
       " 'nasnetalarge',\n",
       " 'nest_base',\n",
       " 'nest_base_jx',\n",
       " 'nest_small',\n",
       " 'nest_small_jx',\n",
       " 'nest_tiny',\n",
       " 'nest_tiny_jx',\n",
       " 'nextvit_base',\n",
       " 'nextvit_large',\n",
       " 'nextvit_small',\n",
       " 'nf_ecaresnet26',\n",
       " 'nf_ecaresnet50',\n",
       " 'nf_ecaresnet101',\n",
       " 'nf_regnet_b0',\n",
       " 'nf_regnet_b1',\n",
       " 'nf_regnet_b2',\n",
       " 'nf_regnet_b3',\n",
       " 'nf_regnet_b4',\n",
       " 'nf_regnet_b5',\n",
       " 'nf_resnet26',\n",
       " 'nf_resnet50',\n",
       " 'nf_resnet101',\n",
       " 'nf_seresnet26',\n",
       " 'nf_seresnet50',\n",
       " 'nf_seresnet101',\n",
       " 'nfnet_f0',\n",
       " 'nfnet_f1',\n",
       " 'nfnet_f2',\n",
       " 'nfnet_f3',\n",
       " 'nfnet_f4',\n",
       " 'nfnet_f5',\n",
       " 'nfnet_f6',\n",
       " 'nfnet_f7',\n",
       " 'nfnet_l0',\n",
       " 'pit_b_224',\n",
       " 'pit_b_distilled_224',\n",
       " 'pit_s_224',\n",
       " 'pit_s_distilled_224',\n",
       " 'pit_ti_224',\n",
       " 'pit_ti_distilled_224',\n",
       " 'pit_xs_224',\n",
       " 'pit_xs_distilled_224',\n",
       " 'pnasnet5large',\n",
       " 'poolformer_m36',\n",
       " 'poolformer_m48',\n",
       " 'poolformer_s12',\n",
       " 'poolformer_s24',\n",
       " 'poolformer_s36',\n",
       " 'poolformerv2_m36',\n",
       " 'poolformerv2_m48',\n",
       " 'poolformerv2_s12',\n",
       " 'poolformerv2_s24',\n",
       " 'poolformerv2_s36',\n",
       " 'pvt_v2_b0',\n",
       " 'pvt_v2_b1',\n",
       " 'pvt_v2_b2',\n",
       " 'pvt_v2_b2_li',\n",
       " 'pvt_v2_b3',\n",
       " 'pvt_v2_b4',\n",
       " 'pvt_v2_b5',\n",
       " 'regnetv_040',\n",
       " 'regnetv_064',\n",
       " 'regnetx_002',\n",
       " 'regnetx_004',\n",
       " 'regnetx_004_tv',\n",
       " 'regnetx_006',\n",
       " 'regnetx_008',\n",
       " 'regnetx_016',\n",
       " 'regnetx_032',\n",
       " 'regnetx_040',\n",
       " 'regnetx_064',\n",
       " 'regnetx_080',\n",
       " 'regnetx_120',\n",
       " 'regnetx_160',\n",
       " 'regnetx_320',\n",
       " 'regnety_002',\n",
       " 'regnety_004',\n",
       " 'regnety_006',\n",
       " 'regnety_008',\n",
       " 'regnety_008_tv',\n",
       " 'regnety_016',\n",
       " 'regnety_032',\n",
       " 'regnety_040',\n",
       " 'regnety_040_sgn',\n",
       " 'regnety_064',\n",
       " 'regnety_080',\n",
       " 'regnety_080_tv',\n",
       " 'regnety_120',\n",
       " 'regnety_160',\n",
       " 'regnety_320',\n",
       " 'regnety_640',\n",
       " 'regnety_1280',\n",
       " 'regnety_2560',\n",
       " 'regnetz_005',\n",
       " 'regnetz_040',\n",
       " 'regnetz_040_h',\n",
       " 'regnetz_b16',\n",
       " 'regnetz_b16_evos',\n",
       " 'regnetz_c16',\n",
       " 'regnetz_c16_evos',\n",
       " 'regnetz_d8',\n",
       " 'regnetz_d8_evos',\n",
       " 'regnetz_d32',\n",
       " 'regnetz_e8',\n",
       " 'repghostnet_050',\n",
       " 'repghostnet_058',\n",
       " 'repghostnet_080',\n",
       " 'repghostnet_100',\n",
       " 'repghostnet_111',\n",
       " 'repghostnet_130',\n",
       " 'repghostnet_150',\n",
       " 'repghostnet_200',\n",
       " 'repvgg_a0',\n",
       " 'repvgg_a1',\n",
       " 'repvgg_a2',\n",
       " 'repvgg_b0',\n",
       " 'repvgg_b1',\n",
       " 'repvgg_b1g4',\n",
       " 'repvgg_b2',\n",
       " 'repvgg_b2g4',\n",
       " 'repvgg_b3',\n",
       " 'repvgg_b3g4',\n",
       " 'repvgg_d2se',\n",
       " 'repvit_m0_9',\n",
       " 'repvit_m1',\n",
       " 'repvit_m1_0',\n",
       " 'repvit_m1_1',\n",
       " 'repvit_m1_5',\n",
       " 'repvit_m2',\n",
       " 'repvit_m2_3',\n",
       " 'repvit_m3',\n",
       " 'res2net50_14w_8s',\n",
       " 'res2net50_26w_4s',\n",
       " 'res2net50_26w_6s',\n",
       " 'res2net50_26w_8s',\n",
       " 'res2net50_48w_2s',\n",
       " 'res2net50d',\n",
       " 'res2net101_26w_4s',\n",
       " 'res2net101d',\n",
       " 'res2next50',\n",
       " 'resmlp_12_224',\n",
       " 'resmlp_24_224',\n",
       " 'resmlp_36_224',\n",
       " 'resmlp_big_24_224',\n",
       " 'resnest14d',\n",
       " 'resnest26d',\n",
       " 'resnest50d',\n",
       " 'resnest50d_1s4x24d',\n",
       " 'resnest50d_4s2x40d',\n",
       " 'resnest101e',\n",
       " 'resnest200e',\n",
       " 'resnest269e',\n",
       " 'resnet10t',\n",
       " 'resnet14t',\n",
       " 'resnet18',\n",
       " 'resnet18d',\n",
       " 'resnet26',\n",
       " 'resnet26d',\n",
       " 'resnet26t',\n",
       " 'resnet32ts',\n",
       " 'resnet33ts',\n",
       " 'resnet34',\n",
       " 'resnet34d',\n",
       " 'resnet50',\n",
       " 'resnet50_gn',\n",
       " 'resnet50c',\n",
       " 'resnet50d',\n",
       " 'resnet50s',\n",
       " 'resnet50t',\n",
       " 'resnet51q',\n",
       " 'resnet61q',\n",
       " 'resnet101',\n",
       " 'resnet101c',\n",
       " 'resnet101d',\n",
       " 'resnet101s',\n",
       " 'resnet152',\n",
       " 'resnet152c',\n",
       " 'resnet152d',\n",
       " 'resnet152s',\n",
       " 'resnet200',\n",
       " 'resnet200d',\n",
       " 'resnetaa34d',\n",
       " 'resnetaa50',\n",
       " 'resnetaa50d',\n",
       " 'resnetaa101d',\n",
       " 'resnetblur18',\n",
       " 'resnetblur50',\n",
       " 'resnetblur50d',\n",
       " 'resnetblur101d',\n",
       " 'resnetrs50',\n",
       " 'resnetrs101',\n",
       " 'resnetrs152',\n",
       " 'resnetrs200',\n",
       " 'resnetrs270',\n",
       " 'resnetrs350',\n",
       " 'resnetrs420',\n",
       " 'resnetv2_50',\n",
       " 'resnetv2_50d',\n",
       " 'resnetv2_50d_evos',\n",
       " 'resnetv2_50d_frn',\n",
       " 'resnetv2_50d_gn',\n",
       " 'resnetv2_50t',\n",
       " 'resnetv2_50x1_bit',\n",
       " 'resnetv2_50x3_bit',\n",
       " 'resnetv2_101',\n",
       " 'resnetv2_101d',\n",
       " 'resnetv2_101x1_bit',\n",
       " 'resnetv2_101x3_bit',\n",
       " 'resnetv2_152',\n",
       " 'resnetv2_152d',\n",
       " 'resnetv2_152x2_bit',\n",
       " 'resnetv2_152x4_bit',\n",
       " 'resnext26ts',\n",
       " 'resnext50_32x4d',\n",
       " 'resnext50d_32x4d',\n",
       " 'resnext101_32x4d',\n",
       " 'resnext101_32x8d',\n",
       " 'resnext101_32x16d',\n",
       " 'resnext101_32x32d',\n",
       " 'resnext101_64x4d',\n",
       " 'rexnet_100',\n",
       " 'rexnet_130',\n",
       " 'rexnet_150',\n",
       " 'rexnet_200',\n",
       " 'rexnet_300',\n",
       " 'rexnetr_100',\n",
       " 'rexnetr_130',\n",
       " 'rexnetr_150',\n",
       " 'rexnetr_200',\n",
       " 'rexnetr_300',\n",
       " 'samvit_base_patch16',\n",
       " 'samvit_base_patch16_224',\n",
       " 'samvit_huge_patch16',\n",
       " 'samvit_large_patch16',\n",
       " 'sebotnet33ts_256',\n",
       " 'sedarknet21',\n",
       " 'sehalonet33ts',\n",
       " 'selecsls42',\n",
       " 'selecsls42b',\n",
       " 'selecsls60',\n",
       " 'selecsls60b',\n",
       " 'selecsls84',\n",
       " 'semnasnet_050',\n",
       " 'semnasnet_075',\n",
       " 'semnasnet_100',\n",
       " 'semnasnet_140',\n",
       " 'senet154',\n",
       " 'sequencer2d_l',\n",
       " 'sequencer2d_m',\n",
       " 'sequencer2d_s',\n",
       " 'seresnet18',\n",
       " 'seresnet33ts',\n",
       " 'seresnet34',\n",
       " 'seresnet50',\n",
       " 'seresnet50t',\n",
       " 'seresnet101',\n",
       " 'seresnet152',\n",
       " 'seresnet152d',\n",
       " 'seresnet200d',\n",
       " 'seresnet269d',\n",
       " 'seresnetaa50d',\n",
       " 'seresnext26d_32x4d',\n",
       " 'seresnext26t_32x4d',\n",
       " 'seresnext26ts',\n",
       " 'seresnext50_32x4d',\n",
       " 'seresnext101_32x4d',\n",
       " 'seresnext101_32x8d',\n",
       " 'seresnext101_64x4d',\n",
       " 'seresnext101d_32x8d',\n",
       " 'seresnextaa101d_32x8d',\n",
       " 'seresnextaa201d_32x8d',\n",
       " 'skresnet18',\n",
       " 'skresnet34',\n",
       " 'skresnet50',\n",
       " 'skresnet50d',\n",
       " 'skresnext50_32x4d',\n",
       " 'spnasnet_100',\n",
       " 'swin_base_patch4_window7_224',\n",
       " 'swin_base_patch4_window12_384',\n",
       " 'swin_large_patch4_window7_224',\n",
       " 'swin_large_patch4_window12_384',\n",
       " 'swin_s3_base_224',\n",
       " 'swin_s3_small_224',\n",
       " 'swin_s3_tiny_224',\n",
       " 'swin_small_patch4_window7_224',\n",
       " 'swin_tiny_patch4_window7_224',\n",
       " 'swinv2_base_window8_256',\n",
       " 'swinv2_base_window12_192',\n",
       " 'swinv2_base_window12to16_192to256',\n",
       " 'swinv2_base_window12to24_192to384',\n",
       " 'swinv2_base_window16_256',\n",
       " 'swinv2_cr_base_224',\n",
       " 'swinv2_cr_base_384',\n",
       " 'swinv2_cr_base_ns_224',\n",
       " 'swinv2_cr_giant_224',\n",
       " 'swinv2_cr_giant_384',\n",
       " 'swinv2_cr_huge_224',\n",
       " 'swinv2_cr_huge_384',\n",
       " 'swinv2_cr_large_224',\n",
       " 'swinv2_cr_large_384',\n",
       " 'swinv2_cr_small_224',\n",
       " 'swinv2_cr_small_384',\n",
       " 'swinv2_cr_small_ns_224',\n",
       " 'swinv2_cr_small_ns_256',\n",
       " 'swinv2_cr_tiny_224',\n",
       " 'swinv2_cr_tiny_384',\n",
       " 'swinv2_cr_tiny_ns_224',\n",
       " 'swinv2_large_window12_192',\n",
       " 'swinv2_large_window12to16_192to256',\n",
       " 'swinv2_large_window12to24_192to384',\n",
       " 'swinv2_small_window8_256',\n",
       " 'swinv2_small_window16_256',\n",
       " 'swinv2_tiny_window8_256',\n",
       " 'swinv2_tiny_window16_256',\n",
       " 'tf_efficientnet_b0',\n",
       " 'tf_efficientnet_b1',\n",
       " 'tf_efficientnet_b2',\n",
       " 'tf_efficientnet_b3',\n",
       " 'tf_efficientnet_b4',\n",
       " 'tf_efficientnet_b5',\n",
       " 'tf_efficientnet_b6',\n",
       " 'tf_efficientnet_b7',\n",
       " 'tf_efficientnet_b8',\n",
       " 'tf_efficientnet_cc_b0_4e',\n",
       " 'tf_efficientnet_cc_b0_8e',\n",
       " 'tf_efficientnet_cc_b1_8e',\n",
       " 'tf_efficientnet_el',\n",
       " 'tf_efficientnet_em',\n",
       " 'tf_efficientnet_es',\n",
       " 'tf_efficientnet_l2',\n",
       " 'tf_efficientnet_lite0',\n",
       " 'tf_efficientnet_lite1',\n",
       " 'tf_efficientnet_lite2',\n",
       " 'tf_efficientnet_lite3',\n",
       " 'tf_efficientnet_lite4',\n",
       " 'tf_efficientnetv2_b0',\n",
       " 'tf_efficientnetv2_b1',\n",
       " 'tf_efficientnetv2_b2',\n",
       " 'tf_efficientnetv2_b3',\n",
       " 'tf_efficientnetv2_l',\n",
       " 'tf_efficientnetv2_m',\n",
       " 'tf_efficientnetv2_s',\n",
       " 'tf_efficientnetv2_xl',\n",
       " 'tf_mixnet_l',\n",
       " 'tf_mixnet_m',\n",
       " 'tf_mixnet_s',\n",
       " 'tf_mobilenetv3_large_075',\n",
       " 'tf_mobilenetv3_large_100',\n",
       " 'tf_mobilenetv3_large_minimal_100',\n",
       " 'tf_mobilenetv3_small_075',\n",
       " 'tf_mobilenetv3_small_100',\n",
       " 'tf_mobilenetv3_small_minimal_100',\n",
       " 'tiny_vit_5m_224',\n",
       " 'tiny_vit_11m_224',\n",
       " 'tiny_vit_21m_224',\n",
       " 'tiny_vit_21m_384',\n",
       " 'tiny_vit_21m_512',\n",
       " 'tinynet_a',\n",
       " 'tinynet_b',\n",
       " 'tinynet_c',\n",
       " 'tinynet_d',\n",
       " 'tinynet_e',\n",
       " 'tnt_b_patch16_224',\n",
       " 'tnt_s_patch16_224',\n",
       " 'tresnet_l',\n",
       " 'tresnet_m',\n",
       " 'tresnet_v2_l',\n",
       " 'tresnet_xl',\n",
       " 'twins_pcpvt_base',\n",
       " 'twins_pcpvt_large',\n",
       " 'twins_pcpvt_small',\n",
       " 'twins_svt_base',\n",
       " 'twins_svt_large',\n",
       " 'twins_svt_small',\n",
       " 'vgg11',\n",
       " 'vgg11_bn',\n",
       " 'vgg13',\n",
       " 'vgg13_bn',\n",
       " 'vgg16',\n",
       " 'vgg16_bn',\n",
       " 'vgg19',\n",
       " 'vgg19_bn',\n",
       " 'visformer_small',\n",
       " 'visformer_tiny',\n",
       " 'vit_base_patch8_224',\n",
       " 'vit_base_patch14_dinov2',\n",
       " 'vit_base_patch14_reg4_dinov2',\n",
       " 'vit_base_patch16_18x2_224',\n",
       " 'vit_base_patch16_224',\n",
       " 'vit_base_patch16_224_miil',\n",
       " 'vit_base_patch16_384',\n",
       " 'vit_base_patch16_clip_224',\n",
       " 'vit_base_patch16_clip_384',\n",
       " 'vit_base_patch16_clip_quickgelu_224',\n",
       " 'vit_base_patch16_gap_224',\n",
       " 'vit_base_patch16_plus_240',\n",
       " 'vit_base_patch16_reg4_gap_256',\n",
       " 'vit_base_patch16_rpn_224',\n",
       " 'vit_base_patch16_siglip_224',\n",
       " 'vit_base_patch16_siglip_256',\n",
       " 'vit_base_patch16_siglip_384',\n",
       " 'vit_base_patch16_siglip_512',\n",
       " 'vit_base_patch16_xp_224',\n",
       " 'vit_base_patch32_224',\n",
       " 'vit_base_patch32_384',\n",
       " 'vit_base_patch32_clip_224',\n",
       " 'vit_base_patch32_clip_256',\n",
       " 'vit_base_patch32_clip_384',\n",
       " 'vit_base_patch32_clip_448',\n",
       " 'vit_base_patch32_clip_quickgelu_224',\n",
       " 'vit_base_patch32_plus_256',\n",
       " 'vit_base_r26_s32_224',\n",
       " 'vit_base_r50_s16_224',\n",
       " 'vit_base_r50_s16_384',\n",
       " 'vit_base_resnet26d_224',\n",
       " 'vit_base_resnet50d_224',\n",
       " 'vit_giant_patch14_224',\n",
       " 'vit_giant_patch14_clip_224',\n",
       " 'vit_giant_patch14_dinov2',\n",
       " 'vit_giant_patch14_reg4_dinov2',\n",
       " 'vit_giant_patch16_gap_224',\n",
       " 'vit_gigantic_patch14_224',\n",
       " 'vit_gigantic_patch14_clip_224',\n",
       " 'vit_huge_patch14_224',\n",
       " 'vit_huge_patch14_clip_224',\n",
       " 'vit_huge_patch14_clip_336',\n",
       " 'vit_huge_patch14_clip_378',\n",
       " 'vit_huge_patch14_clip_quickgelu_224',\n",
       " 'vit_huge_patch14_clip_quickgelu_378',\n",
       " 'vit_huge_patch14_gap_224',\n",
       " 'vit_huge_patch14_xp_224',\n",
       " 'vit_huge_patch16_gap_448',\n",
       " 'vit_large_patch14_224',\n",
       " 'vit_large_patch14_clip_224',\n",
       " 'vit_large_patch14_clip_336',\n",
       " 'vit_large_patch14_clip_quickgelu_224',\n",
       " 'vit_large_patch14_clip_quickgelu_336',\n",
       " 'vit_large_patch14_dinov2',\n",
       " 'vit_large_patch14_reg4_dinov2',\n",
       " 'vit_large_patch14_xp_224',\n",
       " 'vit_large_patch16_224',\n",
       " 'vit_large_patch16_384',\n",
       " 'vit_large_patch16_siglip_256',\n",
       " 'vit_large_patch16_siglip_384',\n",
       " 'vit_large_patch32_224',\n",
       " 'vit_large_patch32_384',\n",
       " 'vit_large_r50_s32_224',\n",
       " 'vit_large_r50_s32_384',\n",
       " 'vit_medium_patch16_gap_240',\n",
       " 'vit_medium_patch16_gap_256',\n",
       " 'vit_medium_patch16_gap_384',\n",
       " 'vit_medium_patch16_reg4_256',\n",
       " 'vit_medium_patch16_reg4_gap_256',\n",
       " 'vit_relpos_base_patch16_224',\n",
       " 'vit_relpos_base_patch16_cls_224',\n",
       " 'vit_relpos_base_patch16_clsgap_224',\n",
       " 'vit_relpos_base_patch16_plus_240',\n",
       " 'vit_relpos_base_patch16_rpn_224',\n",
       " 'vit_relpos_base_patch32_plus_rpn_256',\n",
       " 'vit_relpos_medium_patch16_224',\n",
       " 'vit_relpos_medium_patch16_cls_224',\n",
       " 'vit_relpos_medium_patch16_rpn_224',\n",
       " 'vit_relpos_small_patch16_224',\n",
       " 'vit_relpos_small_patch16_rpn_224',\n",
       " 'vit_small_patch8_224',\n",
       " 'vit_small_patch14_dinov2',\n",
       " 'vit_small_patch14_reg4_dinov2',\n",
       " 'vit_small_patch16_18x2_224',\n",
       " 'vit_small_patch16_36x1_224',\n",
       " 'vit_small_patch16_224',\n",
       " 'vit_small_patch16_384',\n",
       " 'vit_small_patch32_224',\n",
       " 'vit_small_patch32_384',\n",
       " 'vit_small_r26_s32_224',\n",
       " 'vit_small_r26_s32_384',\n",
       " 'vit_small_resnet26d_224',\n",
       " 'vit_small_resnet50d_s16_224',\n",
       " 'vit_so150m_patch16_reg4_gap_256',\n",
       " 'vit_so150m_patch16_reg4_map_256',\n",
       " 'vit_so400m_patch14_siglip_224',\n",
       " 'vit_so400m_patch14_siglip_384',\n",
       " 'vit_srelpos_medium_patch16_224',\n",
       " 'vit_srelpos_small_patch16_224',\n",
       " 'vit_tiny_patch16_224',\n",
       " 'vit_tiny_patch16_384',\n",
       " 'vit_tiny_r_s16_p8_224',\n",
       " 'vit_tiny_r_s16_p8_384',\n",
       " 'volo_d1_224',\n",
       " 'volo_d1_384',\n",
       " 'volo_d2_224',\n",
       " 'volo_d2_384',\n",
       " 'volo_d3_224',\n",
       " 'volo_d3_448',\n",
       " 'volo_d4_224',\n",
       " 'volo_d4_448',\n",
       " 'volo_d5_224',\n",
       " 'volo_d5_448',\n",
       " 'volo_d5_512',\n",
       " 'vovnet39a',\n",
       " 'vovnet57a',\n",
       " 'wide_resnet50_2',\n",
       " 'wide_resnet101_2',\n",
       " 'xception41',\n",
       " ...]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# List of Vit Models with Pretrained Weight\n",
    "all_models = timm.list_models('*')\n",
    "all_models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "eb192149-94dd-414f-90f4-860962a8db09",
   "metadata": {},
   "outputs": [],
   "source": [
    "mypretrainedname = 'davit_tiny' #Put one of the model name listed in previous cell"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c80e4a36-72f7-4a30-9ae4-9807a63e2650",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9d672b8c6e24409f9f4f8ef8f6bb9f73",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "pytorch_model.bin:   0%|          | 0.00/114M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\RYZEN\\.conda\\envs\\timm\\lib\\site-packages\\huggingface_hub\\file_download.py:149: UserWarning: `huggingface_hub` cache-system uses symlinks by default to efficiently store duplicated files but your machine does not support them in C:\\Users\\RYZEN\\.cache\\huggingface\\hub\\models--timm--davit_tiny.msft_in1k. Caching files will still work but in a degraded version that might require more space on your disk. This warning can be disabled by setting the `HF_HUB_DISABLE_SYMLINKS_WARNING` environment variable. For more details, see https://huggingface.co/docs/huggingface_hub/how-to-cache#limitations.\n",
      "To support symlinks on Windows, you either need to activate Developer Mode or to run Python as an administrator. In order to see activate developer mode, see this article: https://docs.microsoft.com/en-us/windows/apps/get-started/enable-your-device-for-development\n",
      "  warnings.warn(message)\n"
     ]
    }
   ],
   "source": [
    "# Load pre-trained ViT model with specified dropout rate\n",
    "model = timm.create_model(mypretrainedname, pretrained=True, num_classes=num_classes, drop_rate=dropout_rate)\n",
    "model = model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "7208aaea-c3e1-4643-b7fe-be677aa27f00",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loss function and optimizer\n",
    "criterion = torch.nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "fb4b6ea6-7f16-45d9-aed4-38c93df0e1f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Custom dataset folder: dataset_cut\n",
      "Number of classes: 2 with class names: ['Bad', 'Good']\n",
      "Preset dropout rate: 0.0\n",
      "Training/Testing Ratio: 228/98\n",
      "Learning rate: 0.0001\n"
     ]
    }
   ],
   "source": [
    "# Summary info before training loop\n",
    "print(f\"Custom dataset folder: {dataset_path}\")\n",
    "print(f\"Number of classes: {num_classes} with class names: {class_names}\")\n",
    "print(f\"Preset dropout rate: {dropout_rate}\")\n",
    "print(f\"Training/Testing Ratio: {train_size}/{val_size}\")\n",
    "print(f\"Learning rate: {learning_rate}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "44b1ed7a-7581-4c36-b2c0-ac372a606c90",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Training and validation\n",
    "train_acc_history, val_acc_history = [], []\n",
    "train_loss_history, val_loss_history = [], []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "0f92c2bc-495b-46e5-8ac1-983c2a0e14a4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/40.. Train loss: 0.6889.. Train accuracy: 0.6184.. Val loss: 0.6673.. Val accuracy: 0.6633.. Remaining time: 29.65 minutes\n",
      "Epoch 2/40.. Train loss: 0.5775.. Train accuracy: 0.6711.. Val loss: 0.7699.. Val accuracy: 0.7857.. Remaining time: 28.07 minutes\n",
      "Epoch 3/40.. Train loss: 0.2710.. Train accuracy: 0.8947.. Val loss: 0.0300.. Val accuracy: 1.0000.. Remaining time: 27.10 minutes\n",
      "Epoch 4/40.. Train loss: 0.0079.. Train accuracy: 1.0000.. Val loss: 0.0012.. Val accuracy: 1.0000.. Remaining time: 26.29 minutes\n",
      "Epoch 5/40.. Train loss: 0.0007.. Train accuracy: 1.0000.. Val loss: 0.0006.. Val accuracy: 1.0000.. Remaining time: 25.54 minutes\n",
      "Epoch 6/40.. Train loss: 0.0004.. Train accuracy: 1.0000.. Val loss: 0.0004.. Val accuracy: 1.0000.. Remaining time: 24.79 minutes\n",
      "Epoch 7/40.. Train loss: 0.0003.. Train accuracy: 1.0000.. Val loss: 0.0003.. Val accuracy: 1.0000.. Remaining time: 24.04 minutes\n",
      "Epoch 8/40.. Train loss: 0.0002.. Train accuracy: 1.0000.. Val loss: 0.0002.. Val accuracy: 1.0000.. Remaining time: 23.30 minutes\n",
      "Epoch 9/40.. Train loss: 0.0002.. Train accuracy: 1.0000.. Val loss: 0.0002.. Val accuracy: 1.0000.. Remaining time: 22.57 minutes\n",
      "Epoch 10/40.. Train loss: 0.0002.. Train accuracy: 1.0000.. Val loss: 0.0002.. Val accuracy: 1.0000.. Remaining time: 21.84 minutes\n",
      "Epoch 11/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 21.12 minutes\n",
      "Epoch 12/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 20.40 minutes\n",
      "Epoch 13/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 19.68 minutes\n",
      "Epoch 14/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 18.96 minutes\n",
      "Epoch 15/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 18.23 minutes\n",
      "Epoch 16/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 17.51 minutes\n",
      "Epoch 17/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 16.79 minutes\n",
      "Epoch 18/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 16.06 minutes\n",
      "Epoch 19/40.. Train loss: 0.0001.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 15.34 minutes\n",
      "Epoch 20/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0001.. Val accuracy: 1.0000.. Remaining time: 14.61 minutes\n",
      "Epoch 21/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 13.88 minutes\n",
      "Epoch 22/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 13.15 minutes\n",
      "Epoch 23/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 12.42 minutes\n",
      "Epoch 24/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 11.69 minutes\n",
      "Epoch 25/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 10.96 minutes\n",
      "Epoch 26/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 10.23 minutes\n",
      "Epoch 27/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 9.50 minutes\n",
      "Epoch 28/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 8.77 minutes\n",
      "Epoch 29/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 8.04 minutes\n",
      "Epoch 30/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 7.31 minutes\n",
      "Epoch 31/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 6.58 minutes\n",
      "Epoch 32/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 5.85 minutes\n",
      "Epoch 33/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 5.12 minutes\n",
      "Epoch 34/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 4.39 minutes\n",
      "Epoch 35/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 3.66 minutes\n",
      "Epoch 36/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 2.93 minutes\n",
      "Epoch 37/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 2.20 minutes\n",
      "Epoch 38/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 1.46 minutes\n",
      "Epoch 39/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 0.73 minutes\n",
      "Epoch 40/40.. Train loss: 0.0000.. Train accuracy: 1.0000.. Val loss: 0.0000.. Val accuracy: 1.0000.. Remaining time: 0.00 minutes\n"
     ]
    }
   ],
   "source": [
    "# Continue training start here\n",
    "start_time = time.time()  # Start time of the training\n",
    "\n",
    "for epoch in range(num_epochs):\n",
    "    epoch_start_time = time.time()  # Start time of the current epoch\n",
    "\n",
    "    # [Your existing training and validation code goes here]\n",
    "\n",
    "    # Calculate average losses and accuracy\n",
    "    # [Your existing code for calculating and storing training/validation losses and accuracies]\n",
    "    # Training\n",
    "    model.train()\n",
    "    train_loss, train_correct = 0.0, 0\n",
    "    for inputs, labels in train_loader:\n",
    "        inputs, labels = inputs.to(device), labels.to(device)\n",
    "        optimizer.zero_grad()\n",
    "        outputs = model(inputs)\n",
    "        loss = criterion(outputs, labels)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        train_loss += loss.item() * inputs.size(0)\n",
    "        _, predictions = torch.max(outputs, 1)\n",
    "        train_correct += torch.sum(predictions == labels.data)\n",
    "    \n",
    "    # Validation\n",
    "    model.eval()\n",
    "    val_loss, val_correct = 0.0, 0\n",
    "    with torch.no_grad():\n",
    "        for inputs, labels in val_loader:\n",
    "            inputs, labels = inputs.to(device), labels.to(device)\n",
    "            outputs = model(inputs)\n",
    "            loss = criterion(outputs, labels)\n",
    "            val_loss += loss.item() * inputs.size(0)\n",
    "            _, predictions = torch.max(outputs, 1)\n",
    "            val_correct += torch.sum(predictions == labels.data)\n",
    "    \n",
    "    # Calculate average losses and accuracy\n",
    "    train_loss = train_loss / train_size\n",
    "    train_acc = train_correct.double() / train_size\n",
    "    val_loss = val_loss / val_size\n",
    "    val_acc = val_correct.double() / val_size\n",
    "\n",
    "    # Save history for plotting\n",
    "    train_acc_history.append(train_acc.cpu().numpy())\n",
    "    train_loss_history.append(train_loss)\n",
    "    val_acc_history.append(val_acc.cpu().numpy())\n",
    "    val_loss_history.append(val_loss)\n",
    "\n",
    "    \n",
    "    epoch_time = time.time() - epoch_start_time  # Time taken for the current epoch\n",
    "    total_time = time.time() - start_time  # Total time elapsed since the start of training\n",
    "    avg_time_per_epoch = total_time / (epoch + 1)  # Average time per epoch\n",
    "    remaining_epochs = num_epochs - epoch - 1\n",
    "    estimated_remaining_time = avg_time_per_epoch * remaining_epochs\n",
    "\n",
    "    # Convert estimated remaining time from seconds to minutes\n",
    "    estimated_remaining_time_min = estimated_remaining_time / 60\n",
    "\n",
    "    # Print results every epoch with estimated remaining time\n",
    "    print(f\"Epoch {epoch+1}/{num_epochs}.. \"\n",
    "          f\"Train loss: {train_loss:.4f}.. \"\n",
    "          f\"Train accuracy: {train_acc:.4f}.. \"\n",
    "          f\"Val loss: {val_loss:.4f}.. \"\n",
    "          f\"Val accuracy: {val_acc:.4f}.. \"\n",
    "          f\"Remaining time: {estimated_remaining_time_min:.2f} minutes\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "05ca6d66-bce0-409b-9ef5-71af7337f8b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot trend chart for Accuracy and Loss\n",
    "plt.figure(figsize=(12, 4))\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(train_acc_history, label='Training Accuracy')\n",
    "plt.plot(val_acc_history, label='Validation Accuracy')\n",
    "plt.legend(loc='lower right')\n",
    "plt.title('Training and Validation Accuracy')\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(train_loss_history, label='Training Loss')\n",
    "plt.plot(val_loss_history, label='Validation Loss')\n",
    "plt.legend(loc='upper right')\n",
    "plt.title('Training and Validation Loss')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "99813ee7-49c4-4a22-abbd-b9dd40c3d4e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model saved to dataset_cut\\davit_tiny_20240305_162405.pth\n"
     ]
    }
   ],
   "source": [
    "# Save the model\n",
    "timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')\n",
    "model_save_path = os.path.join(dataset_path, mypretrainedname + f'_{timestamp}.pth')\n",
    "torch.save(model.state_dict(), model_save_path)\n",
    "print(f\"Model saved to {model_save_path}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b0221256-4d67-48f2-aa02-8e23dd836301",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Predict Function\n",
    "def predict_image(image_path, model, device, transform, class_names):\n",
    "    image = Image.open(image_path).convert('RGB')\n",
    "    inputs = transform(image).unsqueeze(0).to(device)\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        outputs = model(inputs)\n",
    "        _, predicted = torch.max(outputs, 1)\n",
    "        confidence = torch.nn.functional.softmax(outputs, dim=1)[0] * 100\n",
    "    _, predicted_confidence = torch.max(confidence, 0)\n",
    "    \n",
    "    plt.imshow(image)\n",
    "    plt.title(f\"{class_names[predicted]} ({confidence[predicted].item():.2f}%)\") #({confidence[predicted]:.2f}%)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "493962e2-1fc6-4aa6-9405-3d044164a848",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<All keys matched successfully>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the model weights \n",
    "filename = model_save_path\n",
    "#filename = dataset_path + \"/\" + \"model_20231220_105848.pth\"\n",
    "\n",
    "loaded_weights = torch.load(filename)\n",
    "\n",
    "# load the weights \n",
    "model.load_state_dict(loaded_weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3d009545-4963-4176-aeac-87beb7abb847",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Bad', 'Good']\n"
     ]
    }
   ],
   "source": [
    "print (class_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "77882b42-6512-4037-a5c6-934f6ba51202",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Your existing predict_image function\n",
    "def predict_image(image_path, model, device, transform, class_names):\n",
    "    image = Image.open(image_path).convert('RGB')\n",
    "    inputs = transform(image).unsqueeze(0).to(device)\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        outputs = model(inputs)\n",
    "        _, predicted = torch.max(outputs, 1)\n",
    "        confidence = torch.nn.functional.softmax(outputs, dim=1)[0] * 100\n",
    "    return class_names[predicted[0]], confidence[predicted[0]].item(), image\n"
   ]
  },
  {
   "cell_type": "raw",
   "id": "3019ecfa-abe8-456a-96f0-a9606001b5f1",
   "metadata": {},
   "source": [
    "# Updated function to process folders and calculate average prediction time\n",
    "def process_folders(image_folder_path, model, device, transform, class_names):\n",
    "    total_images = 0\n",
    "    misclassified_images = 0\n",
    "    total_prediction_time = 0\n",
    "\n",
    "    for class_folder in os.listdir(image_folder_path):\n",
    "        folder_path = os.path.join(image_folder_path, class_folder)\n",
    "        if os.path.isdir(folder_path):\n",
    "            for image_file in os.listdir(folder_path):\n",
    "                total_images += 1\n",
    "                image_path = os.path.join(folder_path, image_file)\n",
    "                \n",
    "                start_time = time.time()\n",
    "                predicted_label, confidence, image = predict_image(image_path, model, device, transform, class_names)\n",
    "                prediction_time = time.time() - start_time\n",
    "                total_prediction_time += prediction_time\n",
    "                \n",
    "                # Count and show image if misclassified\n",
    "                if predicted_label != class_folder:\n",
    "                    misclassified_images += 1\n",
    "                    plt.imshow(image)\n",
    "                    plt.title(f\"Predicted: {predicted_label} ({confidence:.2f}%), Actual: {class_folder}\")\n",
    "                    plt.show()\n",
    "\n",
    "    # Calculate and display results\n",
    "    if total_images > 0:\n",
    "        misclassification_percentage = (misclassified_images / total_images) * 100\n",
    "        average_prediction_time = total_prediction_time / total_images\n",
    "        print(f\"Total Images: {total_images}\")\n",
    "        print(f\"Misclassified Images: {misclassified_images}\")\n",
    "        print(f\"Misclassification Percentage: {misclassification_percentage:.2f}%\")\n",
    "        print(f\"Average Prediction Time per Image: {average_prediction_time:.4f} seconds\")\n",
    "    else:\n",
    "        print(\"No images found in the provided folder.\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9d2920b1-edd0-4cc3-901b-275e9461fb0e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm.notebook import tqdm  # Use tqdm.notebook.tqdm for Jupyter Notebook\n",
    "\n",
    "def process_folders(image_folder_path, model, device, transform, class_names):\n",
    "    total_images = 0\n",
    "    misclassified_images = 0\n",
    "    total_prediction_time = 0\n",
    "    \n",
    "    # Get the list of class folders\n",
    "    class_folders = [f for f in os.listdir(image_folder_path) if os.path.isdir(os.path.join(image_folder_path, f))]\n",
    "    \n",
    "    # Wrap the loop with tqdm for the progress bar\n",
    "    for class_folder in tqdm(class_folders, desc='Processing folders'):\n",
    "        folder_path = os.path.join(image_folder_path, class_folder)\n",
    "        \n",
    "        # Get the list of image files\n",
    "        image_files = [f for f in os.listdir(folder_path) if os.path.isfile(os.path.join(folder_path, f))]\n",
    "        \n",
    "        # Wrap the loop with tqdm for the progress bar\n",
    "        for image_file in tqdm(image_files, desc=f'Processing {class_folder}', leave=False):\n",
    "            total_images += 1\n",
    "            image_path = os.path.join(folder_path, image_file)\n",
    "            \n",
    "            start_time = time.time()\n",
    "            predicted_label, confidence, image = predict_image(image_path, model, device, transform, class_names)\n",
    "            prediction_time = time.time() - start_time\n",
    "            total_prediction_time += prediction_time\n",
    "            \n",
    "            # Count and show image if misclassified\n",
    "            if predicted_label != class_folder:\n",
    "                misclassified_images += 1\n",
    "                plt.imshow(image)\n",
    "                plt.title(f\"Predicted: {predicted_label} ({confidence:.2f}%), Actual: {class_folder}\")\n",
    "                plt.show()\n",
    "\n",
    "    # Calculate and display results\n",
    "    if total_images > 0:\n",
    "        misclassification_percentage = (misclassified_images / total_images) * 100\n",
    "        average_prediction_time = total_prediction_time / total_images\n",
    "        print(f\"Total Images: {total_images}\")\n",
    "        print(f\"Misclassified Images: {misclassified_images}\")\n",
    "        print(f\"Misclassification Percentage: {misclassification_percentage:.2f}%\")\n",
    "        print(f\"Average Prediction Time per Image: {average_prediction_time:.4f} seconds\")\n",
    "    else:\n",
    "        print(\"No images found in the provided folder.\")\n",
    "\n",
    "# Example usage:\n",
    "# process_folders('path_to_image_folder', model, device, transform, class_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "306d72c6-dd0c-49b3-a6d5-af7c632dee4a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d37f2e5a2121421a824cbedc0622d243",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Processing folders:   0%|          | 0/2 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Processing Bad:   0%|          | 0/218 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Processing Good:   0%|          | 0/108 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Images: 326\n",
      "Misclassified Images: 0\n",
      "Misclassification Percentage: 0.00%\n",
      "Average Prediction Time per Image: 0.0542 seconds\n"
     ]
    }
   ],
   "source": [
    "# Example usage\n",
    "image_folder_path = dataset_path #'path_to_your_image_folder'\n",
    "# model, device, transform, and class_names need to be defined as per your model and dataset\n",
    "process_folders(image_folder_path, model, device, transform, class_names)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bba9f962-c37d-4cfa-b621-f0a2e923e3e3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}