app.py

from flask import Flask, request, jsonify
from llama_cpp import Llama
# from transformers import AutoModelForCausalLM, AutoTokenizer
llm = Llama.from_pretrained(
	repo_id="bartowski/Llama-3.2-1B-Instruct-GGUF",
	filename="Llama-3.2-1B-Instruct-IQ3_M.gguf",
)
# MODEL_NAME = "meta-llama/Llama-3.1-8B-Instruct"

# tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
# model = AutoModelForCausalLM.from_pretrained(MODEL_NAME).to("cpu")

app = Flask(__name__)
# DEFAULT_TEMPERATURE = 0.7
# DEFAULT_MAX_TOKENS = 150
# DEFAULT_TOP_P = 0.95

def generate_journal_suggestion(current_page):
    try:
        suggestion_prompt = (
            f"""Pe baza înregistrării din jurnal: '{current_page}', generează o singură întrebare pe care utilizatorul ar putea să și-o pună într-un jurnal.
            Întrebarea ar trebui să încurajeze reflecția personală mai profundă, explorarea sentimentelor sau clarificarea obiectivelor."""
        )
        
        # input_ids = tokenizer(suggestion_prompt, return_tensors="pt").input_ids.to("cpu")

        # output_ids = model.generate(
        #     input_ids,
        #     max_length=DEFAULT_MAX_TOKENS,
        #     temperature=DEFAULT_TEMPERATURE,
        #     top_p=DEFAULT_TOP_P,
        #     do_sample=True,
        # )
        
        suggestion_response = llm.create_chat_completion(
        	messages = [
        		{
        			"role": "user",
        			"content": suggestion_prompt
        		}
        	]
        )
        return suggestion_response
        # suggestion_response = tokenizer.decode(output_ids[0], skip_special_tokens=True)

    except Exception as e:
        return f"Error: {str(e)}"


@app.route("/", methods=["POST", "GET"])
def home():
    return "Hi!"

@app.route("/chat", methods=["POST"])
def chat():
    data = request.json
    message = data.get("message", "")
    system_message = data.get("system_message", "You are a friendly chatbot.")
    journal_page = data.get("journal_page", "")

    suggestion = ""
    if journal_page:
        suggestion = generate_journal_suggestion(journal_page)

    return jsonify({"journal_suggestion": suggestion})

if __name__ == "__main__":
    app.run(debug=True)