import gradio as gr
from transformers import pipeline

# Load the model pipeline
pipe = pipeline("image-classification", "dima806/medicinal_plants_image_detection")

# Define the image classification function
def image_classifier(image):
    # Perform image classification
    outputs = pipe(image)
    results = {}
    for result in outputs:
        results[result['label']] = result['score']
    return results

# Define app title and description with HTML formatting
title = "<h1 style='text-align: center; color: #4CAF50;'>Image Classification</h1>"
description = "<p style='text-align: center; font-size: 18px;'>This application serves to classify skin lesion images based on their skin cancer type. Trained using Vision Transformer (ViT), it has achieved a validation accuracy of 86%.</p>"

# Define custom CSS styles for the Gradio app
custom_css = """
            .gradio-interface {
                max-width: 600px;
                margin: auto;
                border-radius: 10px;
                box-shadow: 0px 0px 10px rgba(0, 0, 0, 0.1);
            }
            .title-container {
                padding: 20px;
                background-color: #f0f0f0;
                border-top-left-radius: 10px;
                border-top-right-radius: 10px;
            }
            .description-container {
                padding: 20px;
            }
            """

# Launch the Gradio interface with custom HTML and CSS
demo = gr.Interface(fn=image_classifier, inputs=gr.Image(type="pil"), outputs="label", title=title, description=description, 
                     theme="gstaff/sketch", css=custom_css,
                   )
demo.launch()





















































# import torch
# from torchvision import transforms
# from PIL import Image
# from torchvision import models
# import gradio.inputs as gi
# import gradio.outputs as go
# import gradio as gr

# # Define the ResNet50 model
# class ResNet50(torch.nn.Module):
#     def __init__(self):
#         super(ResNet50, self).__init__()
#         self.resnet = models.resnet50(pretrained=True)
#         for param in self.resnet.parameters():
#             param.requires_grad = False
#         self.resnet.fc = torch.nn.Sequential(
#             torch.nn.Linear(2048, 2)
#         )

#     def forward(self, x):
#         x = self.resnet(x)
#         return x

# # Load the pre-trained model
# model = ResNet50()
# model.load_state_dict(torch.load('best_modelv2.pth', map_location=torch.device('cpu')))
# model.eval()

# # Define transform for input images
# data_transforms = transforms.Compose([
#     transforms.Resize((224, 224)),
#     transforms.ToTensor(),
#     transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
# ])

# # Function to predict image label
# def predict_image_label(image):
#     # Preprocess the image
#     image = data_transforms(image).unsqueeze(0)

#     # Make prediction
#     with torch.no_grad():
#         output = model(image)
#         _, predicted = torch.max(output, 1)

#     label = 'Leaf' if predicted.item() == 0 else 'Plant'
#     return label

# # Create Gradio interface
# # image = gi.Image(shape=(224, 224))
# label = go.Label(num_top_classes=2)

# gr.Interface(fn=predict_image_label,inputs="image", outputs=label, title="Leaf or Plant Classifier").launch()