ky32/Email_Classification

1from fastapi import FastAPI, HTTPException2from transformers import AutoModelForCausalLM, AutoTokenizer3import os4import copy5import time6import llama_cpp7from llama_cpp import Llama8from huggingface_hub import hf_hub_download  9 10import transformers11import torch12 13 14app = FastAPI()15 16"""17model_path = hf_hub_download(18        repo_id="TheBloke/Mistral-7B-v0.1-GGUF",19        filename="mistral-7b-v0.1.Q4_K_M.gguf")20 21llm = Llama(22    model_path=model_path,23    n_ctx=2048,24    n_threads=225) 26"""27 28@app.get("/")29async def generate_text():30    try:31        """32        output = llm(33          "Q: Name the planets in the solar system? A: ", 34          max_tokens=32, 35          stop=["Q:", "\n"], 36          echo=True)37        output = llm.create_chat_completion(38        messages=[39        {40            "role": "system",41            "content": "You are a helpful assistant that outputs in JSON.",42        },43        {"role": "user", "content": "Who won the world series in 2020"},44    ],45    response_format={46        "type": "json_object",47    },48    temperature=0.7,49)50    """51    model_id = "meta-llama/Meta-Llama-3-8B"52    pipeline = transformers.pipeline("text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device_map="auto")53    pipeline("Hey how are you doing today?")54        return  pipeline("Hey how are you doing today?")55    except Exception as e:56        raise HTTPException(status_code=500, detail=str(e))