Add files via upload

still progressing

Author: qkrwoghd04

Image_preprocess.py

@@ -0,0 +1,38 @@
+# Image Classification
+import torch
+from torchvision.transforms import v2
+# Visualization
+import matplotlib.pyplot as plt
+
+def image_transform():
+
+  H, W = 224, 224
+  img = torch.randint(0, 256, size=(3, H, W), dtype=torch.uint8)
+
+  transforms = v2.Compose([
+      v2.RandomResizedCrop(size=(224, 224), antialias=True),
+      v2.RandomPhotometricDistort(p=1),
+      v2.RandomChannelPermutation() ,# 채널 무작위 변경
+      v2.RandomHorizontalFlip(p=0.2),
+      v2.ToDtype(torch.float32, scale=True),
+      v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+  ])
+ 
+
+  return transforms
+
+
+def visualize_data(train_dataloader):
+    # 이미지와 정답(label)을 표시합니다.
+    train_features, train_labels = next(iter(train_dataloader))
+    print(f"Feature batch shape: {train_features.size()}")
+    print(f"Labels batch shape: {train_labels.size()}")
+    img = train_features[0].squeeze()
+    label = train_labels[0]
+
+    if img.dim() == 3 and img.size(0) == 3:
+        img = img.permute(1, 2, 0)
+        
+    plt.imshow(img, cmap="gray")
+    plt.show()
+    print(f"Label: {label}")

custom_image_dataset.py

@@ -0,0 +1,31 @@
+import os
+import pandas as pd
+from torch.utils.data import Dataset
+from torchvision.io import read_image
+
+
+class CustomImageDataset():
+  def __init__(self, csv_file, img_dir, transform=None, target_transform=None):
+        self.img_file = pd.read_csv(os.path.join(img_dir, csv_file))
+        self.img_dir = img_dir
+        self.transform = transform
+        self.target_transform = target_transform
+    
+  def __len__(self):
+    return len(self.img_file)
+    
+  def __getitem__(self, index):
+    # 레이블을 기반으로 서브폴더 경로 결정
+    label = self.img_file.iloc[index, 2]
+    subfolder = 'Sleep' if label == 'Sleep' else 'Fall'
+    img_path = os.path.join(self.img_dir, 'train', subfolder, self.img_file.iloc[index, 0])
+    if(label == "Sleep"):
+       label = 0
+    else: label = 1
+    image = read_image(img_path)
+    if self.transform:
+        image = self.transform(image)
+    if self.target_transform:
+        label = self.target_transform(label)
+
+    return image, label

vit_model.py

@@ -0,0 +1,77 @@
+# Pytorch
+import sys
+import torch
+
+print(torch.__version__)
+from torch import nn
+import os
+from torch.utils.data import DataLoader
+from torchvision import datasets
+from custom_image_dataset import CustomImageDataset
+import argparse
+from Image_preprocess import *
+
+device = (
+    "cuda"
+    if torch.cuda.is_available()
+    else "mps" if torch.backends.mps.is_available() else "cpu"
+)
+print(f"Using {device} device")
+
+
+class NerualNetwork(nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.flatten = nn.Flatten()
+        self.linear_relu_stack = nn.Sequential(
+            nn.Linear(224 * 224, 512),
+            nn.RuLU(),
+            nn.Linear(512, 512),
+            nn.RuLU(),
+            nn.Linear(512, 2),
+        )
+
+    def forward(self, x):
+        x = self.flatten(x)
+        logits = self.linear_relu_stack(x)
+        return logits
+
+
+def main(argv):
+
+    parser = argparse.ArgumentParser(
+        description="Fall detection with Vision Transformer(ViT) Model",
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
+    )
+    parser.add_argument(
+        "-l",
+        "--load",
+        nargs="+",
+        type=str,
+        help="python vit_model.py --load train_captions.csv",
+    )
+    args = parser.parse_args()
+
+    ## data load & argumentation
+    if args.load is not None:
+        file_path = "C:/Users/Jaeho/OneDrive/바탕 화면/fall detection/dataset/"
+        data_csv = args.load[0]
+        print("Loading file...")
+        transform = image_transform()
+        train_dataset = CustomImageDataset(
+            csv_file=data_csv, img_dir=file_path, transform=transform
+        )
+        # test_dataset = CustomImageDataset(data_csv, file_path, transform)
+        train_dataloader = torch.utils.data.DataLoader(
+            train_dataset, batch_size=1024, shuffle=True, num_workers=4
+        )
+        # test_dataloader = torch.utils.data.DataLoader(test_dataset,
+        #                                           batch_size=1024,
+        #                                           shuffle=True,
+        #                                           num_workers=4)
+
+        visualize_data(train_dataloader)
+
+
+if __name__ == "__main__":
+    main(sys.argv[1:])