gemma.py

"""
#git+https://github.com/huggingface/transformers
transformers==4.43.1
huggingface_hub
bitsandbytes
accelerate
langchain
torch
flask 
gunicorn
twilio
baseten 
spaces
"""




from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
from flask import Flask, request, jsonify
import torch
import os

app = Flask(__name__)

print("Hello welcome to Sema AI", flush=True)  # Flush to ensure immediate output

@app.route("/")
def hello():
    return "hello 🤗, Welcome to Sema AI Chat Service."

# Get Hugging Face credentials from environment variables
HF_TOKEN = os.getenv('HF_TOKEN')

if not HF_TOKEN:
    print("Missing Hugging Face token", flush=True)

model_id = "google/gemma-2-2b-it"
device = "cuda:0" if torch.cuda.is_available() else "cpu"

# Load tokenizer and model with authentication token
tokenizer = AutoTokenizer.from_pretrained(model_id, token=HF_TOKEN)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype=torch.float16,
    token=HF_TOKEN
)

app_pipeline = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer
)

@app.route("/generate_text", methods=["POST"])
def generate_text():
    data = request.json
    prompt = data.get("prompt", "")
    max_new_tokens = data.get("max_new_tokens", 1000)
    do_sample = data.get("do_sample", True)
    temperature = data.get("temperature", 0.1)
    top_k = data.get("top_k", 50)
    top_p = data.get("top_p", 0.95)
    print(f"{prompt}: ")

    try:
        outputs = app_pipeline(
            prompt,
            max_new_tokens=max_new_tokens,
            do_sample=do_sample,
            temperature=temperature,
            top_k=top_k,
            top_p=top_p
        )
        response_text = outputs[0]["generated_text"]
    except Exception as e:
        return jsonify({"error": str(e)}), 500

    return jsonify({"response": response_text})

if __name__ == "__main__":
    app.run(debug=False, port=8888)