Update app.py

app.py

@@ -1,17 +1,4 @@
-import os
-import zipfile
-
-zip_file = "Fruits_datasets.zip"
-
-# Extract the zip file
-with zipfile.ZipFile(zip_file, 'r') as zip_ref:
-    zip_ref.extractall(".")
-    print("Extracted files:")
-    for name in zip_ref.namelist():
-        print(name)
-
 
-"""
 import os
 import zipfile
 import torch
@@ -29,11 +16,15 @@ import torchvision
 
 
 
+import os
+import zipfile
+from PIL import Image
+
 zip_path = "Fruits_dataset.zip"
 extract_root = "./unzipped"
 dataset_folder = None
 
-# Extract ZIP
+# Step 1: Extract ZIP
 if not os.path.exists(extract_root):
     print("Extracting dataset from uploaded zip...")
     with zipfile.ZipFile(zip_path, 'r') as zip_ref:
@@ -41,21 +32,24 @@ if not os.path.exists(extract_root):
 else:
     print("Dataset already extracted.")
 
-# Try to auto-detect the actual dataset folder
+# Step 2: Auto-detect the dataset folder (looking for class folders like Apples, Pears, etc.)
 for root, dirs, files in os.walk(extract_root):
-    if any(os.path.isdir(os.path.join(root, d)) for d in dirs):
-        # Pick folder that has class folders inside
-        if all(os.path.isdir(os.path.join(root, d)) for d in dirs):
+    # Check if this folder contains class folders (e.g., Apples/)
+    if all(os.path.isdir(os.path.join(root, d)) for d in dirs) and len(dirs) > 0:
+        # Ensure it has image files inside class folders
+        subdir = os.path.join(root, dirs[0])
+        if any(f.lower().endswith(('.png', '.jpg', '.jpeg')) for f in os.listdir(subdir)):
             dataset_folder = root
             break
 
+# Step 3: Handle case when nothing found
 if dataset_folder is None:
-    raise RuntimeError("Could not find dataset folder with class directories.")
+    raise RuntimeError("❌ Could not find dataset folder with class image directories.")
 
-print(f"Detected dataset folder: {dataset_folder}")
-print("Classes:", os.listdir(dataset_folder))
+print(f"✅ Detected dataset folder: {dataset_folder}")
+print("✅ Classes:", os.listdir(dataset_folder))
 
-# Show image sizes
+# Step 4: Show image dimensions per class
 for cls in os.listdir(dataset_folder):
     cls_path = os.path.join(dataset_folder, cls)
     if os.path.isdir(cls_path):
@@ -65,7 +59,7 @@ for cls in os.listdir(dataset_folder):
                 img = Image.open(img_path)
                 print(f"{img_path}: {img.size}")
             except Exception as e:
-                print(f"Failed to open {img_path}: {e}")
+                print(f"⚠️ Failed to open {img_path}: {e}")
 
 # -------------------------------
 # 2. Load Data
@@ -211,6 +205,6 @@ with desc_block:
         image_input = gr.Image(type="pil", label="Upload Fruit Image", image_mode="RGB", height=256, width=256)
         pred_output = gr.Textbox(label="Prediction Result")
         predict_button = gr.Button("Classify Image")
-        predict_button.click(fn=predict_image, inputs=image_input, outputs=pred_output)"""
+        predict_button.click(fn=predict_image, inputs=image_input, outputs=pred_output)
 
 desc_block.launch()