app.py

import torch
from torchvision import models, transforms
from PIL import Image
import gradio as gr

# Updated class names with 'plaque' in front of 'calculus' and 'gingivitis'
class_names = [
    "plaque_calculus",
    "caries",
    "plaque_gingivitis",
    "hypodontia",
    "mouth_ulcer",
    "tooth_discoloration"
]

model = models.resnet50(weights=None)
model.fc = torch.nn.Linear(model.fc.in_features, len(class_names))

model.load_state_dict(torch.load('tooth_model.pth', map_location=torch.device('cpu')))
model.eval()

preprocess = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])

def predict_image(image):
    processed_image = preprocess(image).unsqueeze(0)

    with torch.no_grad():
        outputs = model(processed_image)
        probabilities = torch.nn.functional.softmax(outputs, dim=1)
        top_probs, top_indices = torch.topk(probabilities, 2)
        top_classes = [class_names[idx] for idx in top_indices[0]]

    # Create a result dictionary with class names and probabilities
    result = {top_classes[i]: top_probs[0][i].item() for i in range(2)}

    return result

iface = gr.Interface(
    fn=predict_image,
    inputs=gr.Image(type="pil"),
    outputs="label",
    title="Medical Image Classification",
    description="Upload an image to predict its class with probabilities of top 2 predictions."
)

iface.launch()