app.py

from flask import Flask, request, jsonify
from transformers import AutoModelForCausalLM, AutoTokenizer
import json
import logging
import os
import threading
import time

# Set up logging to stdout only
logging.basicConfig(
    level=logging.DEBUG,
    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
    handlers=[
        logging.StreamHandler()  # Log to stdout
    ]
)
logger = logging.getLogger(__name__)

# Initialize Flask app
app = Flask(__name__)

# Global variables for model and tokenizer
model = None
tokenizer = None
model_load_status = "not_loaded"

# Define model path and fallback
model_path = "/app/fine-tuned-construction-llm"
fallback_model = "distilgpt2"

# Function to load model in the background
def load_model_background():
    global model, tokenizer, model_load_status
    try:
        if os.path.isdir(model_path):
            logger.info(f"Loading local model from {model_path}")
            model = AutoModelForCausalLM.from_pretrained(model_path, local_files_only=True)
            tokenizer = AutoTokenizer.from_pretrained(model_path, local_files_only=True)
            model_load_status = "local_model_loaded"
        else:
            logger.info(f"Model directory not found: {model_path}. Using pre-trained model: {fallback_model}")
            model = AutoModelForCausalLM.from_pretrained(fallback_model)
            tokenizer = AutoTokenizer.from_pretrained(fallback_model)
            model_load_status = "fallback_model_loaded"
        logger.info("Model and tokenizer loaded successfully")
    except Exception as e:
        logger.error(f"Failed to load model or tokenizer: {str(e)}")
        model_load_status = f"failed: {str(e)}"

# Start model loading in a background thread
def start_background_tasks():
    logger.debug("Starting background tasks")
    thread = threading.Thread(target=load_model_background)
    thread.daemon = True
    thread.start()

@app.route("/")
def root():
    logger.debug("Root endpoint accessed")
    return jsonify({"message": "Supervisor AI Coach is running"})

@app.route("/health")
def health_check():
    logger.debug("Health endpoint accessed")
    return jsonify({
        "status": "healthy" if model_load_status in ["local_model_loaded", "fallback_model_loaded"] else "starting",
        "model_load_status": model_load_status
    })

@app.route("/generate_coaching", methods=["POST"])
def generate_coaching():
    logger.debug("Generate coaching endpoint accessed")
    # Manual validation of request data (replacing Pydantic)
    data = request.get_json()
    if not data:
        logger.error("Invalid request: No JSON data provided")
        return jsonify({"error": "Invalid request: JSON data required"}), 400

    required_fields = ["role", "project_id", "milestones", "reflection_log"]
    missing_fields = [field for field in required_fields if field not in data]
    if missing_fields:
        logger.error(f"Missing required fields: {missing_fields}")
        return jsonify({"error": f"Missing required fields: {missing_fields}"}), 400

    if model is None or tokenizer is None:
        logger.warning("Model or tokenizer not loaded")
        # Return a static response if the model isn't loaded yet
        response_json = {
            "checklist": ["Inspect safety equipment", "Review milestone progress"],
            "tips": ["Prioritize team communication", "Check weather updates"],
            "quote": "Every step forward counts!"
        }
        return jsonify(response_json)

    try:
        # Prepare input text
        input_text = (
            f"Role: {data['role']}, Project: {data['project_id']}, "
            f"Milestones: {data['milestones']}, Reflection: {data['reflection_log']}"
        )
        
        # Tokenize input
        inputs = tokenizer(input_text, return_tensors="pt", max_length=512, truncation=True)
        
        # Generate output
        outputs = model.generate(
            inputs["input_ids"],
            max_length=200,
            num_return_sequences=1,
            no_repeat_ngram_size=2,
            do_sample=True,
            temperature=0.7
        )
        
        # Decode and parse response
        response_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
        
        # Since distilgpt2 may not output JSON, parse the response manually or use fallback
        if not response_text.startswith("{"):
            checklist = ["Inspect safety equipment", "Review milestone progress"]
            tips = ["Prioritize team communication", "Check weather updates"]
            quote = "Every step forward counts!"
            response_json = {"checklist": checklist, "tips": tips, "quote": quote}
            logger.warning("Model output is not JSON, using default response")
        else:
            try:
                response_json = json.loads(response_text)
            except json.JSONDecodeError:
                response_json = {
                    "checklist": ["Inspect safety equipment", "Review milestone progress"],
                    "tips": ["Prioritize team communication", "Check weather updates"],
                    "quote": "Every step forward counts!"
                }
                logger.warning("Failed to parse model output as JSON, using default response")
        
        return jsonify(response_json)
    
    except Exception as e:
        logger.error(f"Error generating coaching response: {str(e)}")
        return jsonify({"error": f"Internal server error: {str(e)}"}), 500

if __name__ == "__main__":
    # Start background tasks before the app runs
    start_background_tasks()
    # Run Flask app with waitress for production-ready WSGI server
    from waitress import serve
    logger.debug("Starting Flask app with Waitress")
    serve(app, host="0.0.0.0", port=7860)