pip install transformers

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
import gradio as gr

# Use a pipeline as a high-level helper
pipe = pipeline("text-generation", model="meta-llama/Meta-Llama-3.1-70B")

# Load model directly
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3.1-70B")
model = AutoModelForCausalLM.from_pretrained("meta-llama/Meta-Llama-3.1-70B")

# Load sentiment analysis pipeline
sentiment_analyzer = pipeline("sentiment-analysis")

# Initialize conversation context
context = []

def predict(context, input_text):
    """Generate response based on context and input."""
    context.append(input_text)
    inputs = tokenizer(" ".join(context), return_tensors="pt")
    outputs = model.generate(inputs.input_ids, max_length=200, pad_token_id=tokenizer.eos_token_id)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    context.append(response)
    return response

def predict_with_emotion(context, input_text):
    """Generate response with emotion detection."""
    sentiment = sentiment_analyzer(input_text)[0]['label']
    response = predict(context, input_text)
    if sentiment == 'NEGATIVE':
        response = "I'm sorry to hear that. " + response
    elif sentiment == 'POSITIVE':
        response = "That's great! " + response
    return response

def chatbot(input_text):
    """Gradio chatbot function."""
    global context
    response = predict_with_emotion(context, input_text)
    return response

# Create Gradio interface
iface = gr.Interface(fn=chatbot, inputs="text", outputs="text", title="Contextual Emotion-Aware LLaMA-70B Chatbot")
iface.launch()