diff --git a/api/main.py b/api/main.py index c341268c623b5a75626764d069859583696673f6..91661131faac946534bd67e478c80726ace517c0 100644 --- a/api/main.py +++ b/api/main.py @@ -1,11 +1,12 @@ from contextlib import asynccontextmanager -from api.optimization import MultiProcessOptimizer -from api.routers import runs from fastapi import BackgroundTasks, FastAPI, Request, Response from fastapi.staticfiles import StaticFiles from requests import request as make_request +from api.optimization import MultiProcessOptimizer +from api.routers import runs + # see https://github.com/tiangolo/fastapi/issues/3091#issuecomment-821522932 and https://github.com/encode/starlette/issues/1094#issuecomment-730346075 for heavy-load computation DEBUG = True diff --git a/task.py b/task.py index d062253eeee5e38db3388d5879da7064d3afee4b..abb26c3fb7ca74950dc4e670f920c29bbcd6d8e6 100644 --- a/task.py +++ b/task.py @@ -5,15 +5,13 @@ from functools import lru_cache from statistics import mean from typing import Union +from cli import argument_parser from datasets import Dataset, load_dataset from evaluate import load as load_metric from llama_cpp import LlamaGrammar, deque -from torch.utils import data -from tqdm import tqdm - -from cli import argument_parser from models import Llama2, LLMModel, OpenAI from opt_types import ModelUsage +from tqdm import tqdm from utils import log_calls, logger SYSTEM_MESSAGE = """