Docs: Updated documentation

README.md

@@ -11,16 +11,79 @@ pinned: false
 
 # ResNet50 trained on ImageNet-1K
 
-Model trained on ImageNet-1K with 1000 classes.
+This is a ResNet50 model trained on ImageNet-1K dataset with 1000 classes. The model can classify a wide variety of images into 1000 different categories.
 
-## Model
+## Model Details
 
-`resnet50_imagenet1k.pth`
+- Architecture: ResNet50
+- Dataset: ImageNet-1K
+- Classes: 1000
+- Input Size: 224x224 pixels
+- Model File: `resnet50_imagenet1k.pth`
+- Training Repository: [Link](https://github.com/pradeep6kumar/ImageNet_v4)
 
-## Usage
+## Quick Start
 
-1. Download the model from the link above.
-2. Use the model in your project.
+1. Clone the repository:
+```bash
+git clone https://huggingface.co/spaces/Shilpaj/ImageNet
+cd ImageNet
+```
+
+2. Download the model:
+```bash
+# Option 1: Using wget
+wget https://huggingface.co/spaces/Shilpaj/ImageNet/blob/main/resnet50_imagenet1k.pth
+
+# Option 2: Manual download
+Download from: https://huggingface.co/spaces/Shilpaj/ImageNet/tree/main/resnet50_imagenet1k.pth
+```
+
+3. Install requirements:
+```bash
+pip install -r requirements.txt
+```
+
+4. Run the demo:
+```bash
+python app.py
+```
+
+## Usage in Your Project
 
 ```python
-```
+from inference import ImageNetClassifier
+
+# Initialize the classifier
+classifier = ImageNetClassifier('resnet50_imagenet1k.pth')
+
+# Classify an image
+image_path = 'path/to/your/image.jpg'
+prediction, confidence = classifier.predict(image_path)
+print(f"Prediction: {prediction}")
+print(f"Confidence: {confidence:.2f}%")
+```
+
+## Example Images
+
+The `assets/examples` directory contains sample images for testing:
+- Bird
+- Car
+- Cat
+- Dog
+- Frog
+- Horse
+- Plane
+- Ship
+- Truck
+
+## Repository Structure
+
+```
+.
+├── app.py              # Gradio web interface
+├── inference.py        # Model inference code
+├── requirements.txt    # Python dependencies
+└── assets/
+    └── examples/      # Example images for testing
+```

app.py

@@ -114,14 +114,26 @@ def main():
         with gr.Blocks() as demo:
             gr.Markdown(
                 """
-                # ImageNet-1K trained on ResNet50v2
+                # ResNet50 trained on ImageNet-1K
+                A large-scale image classification dataset with 1.2 million training images across 1,000 object categories.
                 """
             )
 
-            with gr.Tab("GradCam"):
+            with gr.Tab("Predictions & GradCAM"):
                 gr.Markdown(
                     """
-                    Visualize Class Activations Maps generated by the model's layer for the predicted class.
+                    View model predictions and visualize where the model is looking using GradCAM.
+                    
+                    ## Steps to use:
+                    1. Upload an image or select one from the examples below
+                    2. Adjust the sliders (optional):
+                       - Activation Map Transparency: Controls the blend between original image and activation map
+                       - Number of Top Predictions: How many top class predictions to show
+                       - Target Layer Number: Which network layer to visualize (deeper layers show higher-level features)
+                    3. Click "Generate GradCAM" to run the model
+                    4. View the results:
+                       - Left: Original uploaded image
+                       - Right: Model predictions and GradCAM visualization showing where the model focused
                     """
                 )
                 
@@ -183,12 +195,6 @@ def main():
 
                 # Examples section for Gradio 5.x
                 examples = [
-                    # [
-                    #     "assets/examples/dog.jpg", 
-                    #     0.5,  # alpha slider
-                    #     3,    # top_k slider
-                    #     4     # target_layer slider
-                    # ],
                     [
                         "assets/examples/cat.jpg",
                         0.5,
@@ -207,12 +213,6 @@ def main():
                         3,
                         4
                     ],
-                    # [
-                    #     "assets/examples/shark-plane.jpg",
-                    #     0.5,
-                    #     3,
-                    #     4
-                    # ],
                     [
                         "assets/examples/car.jpg",
                         0.5,