Delete app.py

app.py

@@ -1,87 +0,0 @@
-import gradio as gr
-import torch.nn.functional as F
-import torchvision.transforms as transforms
-from huggingface_hub import hf_hub_download
-from PIL import Image
-import torch
-import os
-
-from models.resnet import SupCEResNet
-
-CLASS_NAMES = [
-    "T-shirt", "Shirt", "Knitwear", "Chiffon", "Sweater", "Hoodie",
-    "Windbreaker", "Jacket", "Down Coat", "Suit", "Shawl", "Dress",
-    "Vest", "Underwear"
-]
-
-# Image normalization (matching Clothing1M training)
-mean = (0.485, 0.456, 0.406)
-std = (0.229, 0.224, 0.225)
-
-# Define preprocessing for user-uploaded images
-transform_input_image = transforms.Compose([
-    transforms.Resize(256),
-    transforms.CenterCrop(224),
-    transforms.ToTensor(),
-    transforms.Normalize(mean, std),
-])
-
-
-def load_model_from_huggingface():
-    model_filename = "ckpt_clothing_resnet50.pth"
-    model_path = hf_hub_download(repo_id="austin-carnahan/c2d-clothing1M", filename=model_filename)
-
-    # Load checkpoint and check keys
-    checkpoint = torch.load(model_path, map_location=torch.device("cpu"), weights_only=False)
-
-
-    # Ensure the model state_dict is loaded properly
-    if "model" in checkpoint:
-        sd = {k.replace("module.", ""): v for k, v in checkpoint["model"].items()}
-    else:
-        sd = checkpoint  # Try loading directly
-
-    model = SupCEResNet("resnet50", num_classes=14, pool=True)
-    model.load_state_dict(sd, strict=False)
-    model.eval()
-
-    print("✅ Model loaded successfully!")
-    return model
-
-model = load_model_from_huggingface()
-
-model = load_model_from_huggingface()
-print("✅ Model loaded successfully from Hugging Face Hub!")
-
-
-def gradio_inference(image):
-    """Process uploaded image, run inference, and return predicted class + confidence."""
-    
-    image = image.convert("RGB")
-    input_tensor = transform_input_image(image).unsqueeze(0)  # Add batch dimension
-
-    # Run inference
-    with torch.no_grad():
-        output = model(input_tensor)
-        probabilities = F.softmax(output, dim=1)  # Convert logits to probabilities
-
-        predicted_class = torch.argmax(probabilities, dim=1).item()
-        confidence = probabilities[0, predicted_class].item()
-
-        # Get human-readable class name
-        class_name = CLASS_NAMES[predicted_class]
-
-    return f"Predicted: {class_name} ({confidence:.2%} confidence)"
-
-# Define Gradio UI
-interface = gr.Interface(
-    fn=gradio_inference, 
-    inputs=gr.Image(type="pil"), 
-    outputs="text",
-    title="Clothing Image Classifier",
-    description="Upload an image of clothing and get a predicted class with confidence score.",
-    examples=["images/tshirt.jpg", "images/pants.jpg"]
-)
-
-# Launch the interface
-interface.launch()