app.py

# Import required libraries
import gradio as gr
from transformers import AutoModelForImageClassification, AutoFeatureExtractor
from PIL import Image
import torch

# Load the pre-trained Vision Transformer model and feature extractor
model = AutoModelForImageClassification.from_pretrained("google/vit-base-patch16-224")
extractor = AutoFeatureExtractor.from_pretrained("google/vit-base-patch16-224")

# Define the function to classify the crop using satellite images
def classify_image(image):
    # Preprocess the image using the feature extractor
    inputs = extractor(images=image, return_tensors="pt")
    
    # Make the prediction using the model
    with torch.no_grad():
        outputs = model(**inputs)
    
    # Extract predicted class
    logits = outputs.logits
    predicted_class_idx = logits.argmax(-1).item()
    
    # Return the predicted class
    return f"Predicted Class: {model.config.id2label[predicted_class_idx]}"

# Define the chatbot function for interaction
def respond(message, chat_history):
    # Respond to the user's question
    response = f"Received your message: '{message}' but currently, I'm focusing on classifying crops!"
    return response

# Create a Gradio interface for crop classification
iface = gr.Interface(
    fn=classify_image, 
    inputs=gr.Image(type="pil"), 
    outputs="text",
    title="Real-Time Crop Classification",
    description="Upload a satellite/mobile image to classify crops.",
)

# Create a Gradio Chatbot for interaction
chatbot = gr.ChatInterface(
    fn=respond,
    title="Agricultural Assistant Chatbot",
    description="Ask questions about crop classification and more."
)

# Combine the crop classifier and chatbot in one interface
with gr.Blocks() as demo:
    gr.Markdown("## Farmovation Chatbot")
    with gr.Tab("Crop Classification"):
        iface.render()
    with gr.Tab("Agriculture Chatbot"):
        chatbot.render()

# Launch the Gradio app in Google Colab
demo.launch()