Update app.py

app.py

@@ -10,10 +10,9 @@ device = torch.device('cpu')
 # Define ResNet-50 Architecture
 model = models.resnet50(weights=None)
 
-# revised full connected layer to 37 (num_classes)
+# Revise fully connected layer to output 37 classes (num_classes = 37)
 model.fc = torch.nn.Linear(2048, 37)
 
-# Load Model weights
 model.load_state_dict(torch.load('./resnet50_model_weights.pth', map_location=device))
 
 model.eval()
@@ -24,6 +23,7 @@ transform = transforms.Compose([
     transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
 ])
 
+# List of class names (37 dog and cat breeds)
 class_names = ['Abyssinian (阿比西尼亞貓)', 'American Bulldog (美國鬥牛犬)', 'American Pit Bull Terrier (美國比特鬥牛梗)',
                'Basset Hound (巴吉度獵犬)', 'Beagle (米格魯)', 'Bengal (孟加拉貓)', 'Birman (緬甸貓)', 'Bombay (孟買貓)', 
                'Boxer (拳師犬)', 'British Shorthair (英國短毛貓)', 'Chihuahua (吉娃娃)', 'Egyptian Mau (埃及貓)', 
@@ -35,38 +35,48 @@ class_names = ['Abyssinian (阿比西尼亞貓)', 'American Bulldog (美國鬥
                'Siamese (暹羅貓)', 'Sphynx (無毛貓)', 'Staffordshire Bull Terrier (史塔福郡鬥牛犬)', 
                'Wheaten Terrier (小麥色梗)', 'Yorkshire Terrier (約克夏犬)']
 
-# predict function
+# Prediction function
 def classify_image(image):
-    image = transform(image).unsqueeze(0).to(device)  # make sure prediction on cpu
+    # Apply transformation and add batch dimension
+    image = transform(image).unsqueeze(0).to(device)  
     with torch.no_grad():
+        # Make predictions using the model
         outputs = model(image)
-        probabilities, indices = torch.topk(outputs, k=3)  # top 3 predictions
-        probabilities = torch.nn.functional.softmax(probabilities, dim=1)  
+        # Apply softmax to get probabilities
+        probabilities = torch.nn.functional.softmax(outputs, dim=1)  
+        # Get the top 3 predictions
+        probabilities, indices = torch.topk(probabilities, k=3)  
+        # Return the class names with their corresponding probabilities
         predictions = [(class_names[idx], prob.item()) for idx, prob in zip(indices[0], probabilities[0])]
-        return {class_name: prob for class_name, prob in predictions}
-
+        return {class_name: prob for class_name, prob in predictions}  # Return raw float numbers
 
+# Path to the folder containing example images
 examples_path = './examples'
 
+# Check if the example images folder exists
 if os.path.exists(examples_path):
     print(f"[INFO] Found examples folder at {examples_path}")
 else:
     print(f"[ERROR] Examples folder not found at {examples_path}")
 
-# Gradio Interface
+# Gradio interface
+# Load example images from the folder
 examples = [[examples_path + "/" + img] for img in os.listdir(examples_path)]
 
-# drop down list
-dropdown = gr.Dropdown(choices=class_names, label="Select a breed", type="value")
+# Create dropdown menu for users to see available classes (as reference, no direct connection to prediction)
+dropdown = gr.Dropdown(choices=class_names, label="Recognizable Breeds", type="value")
 
+# Define Gradio Interface
 demo = gr.Interface(
     fn=classify_image,  
-    inputs=[gr.Image(type="pil"), dropdown],  
-    outputs=[gr.Label(num_top_classes=3, label="Top 3 Predictions")],  
+    inputs=[gr.Image(type="pil")],  # Only image input is used for prediction
+    outputs=[gr.Label(num_top_classes=3, label="Top 3 Predictions")],  # Outputs top 3 predictions with probabilities
     examples=examples,  
     title='Oxford Pet 🐈🐕',  
-    description='A ResNet50-based model for classifying 37 different pet breeds.',
-    article='[Oxford Project](https://github.com/Eric-Chung-0511/Learning-Record/tree/main/Data%20Science%20Projects/The%20Oxford-IIIT%20Pet%20Project)'
+    description='A ResNet50-based model for classifying 37 different pet breeds.', 
+    article='[Oxford Project](https://github.com/Eric-Chung-0511/Learning-Record/tree/main/Data%20Science%20Projects/The%20Oxford-IIIT%20Pet%20Project)',  
+    inputs_dropdown=dropdown  # Dropdown used as a reference list
 )
 
+# Launch Gradio demo
 demo.launch()