from fastapi import FastAPI, HTTPException from fastapi.staticfiles import StaticFiles from pydantic import BaseModel import pickle import uvicorn import logging import os import shutil import subprocess import torch from langchain.chains import RetrievalQA from langchain.embeddings import HuggingFaceInstructEmbeddings from langchain.prompts import PromptTemplate # from langchain.embeddings import HuggingFaceEmbeddings from run_localGPT import load_model from prompt_template_utils import get_prompt_template # from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler from langchain.vectorstores import Chroma from werkzeug.utils import secure_filename from constants import CHROMA_SETTINGS, EMBEDDING_MODEL_NAME, PERSIST_DIRECTORY, MODEL_ID, MODEL_BASENAME if torch.backends.mps.is_available(): DEVICE_TYPE = "mps" elif torch.cuda.is_available(): DEVICE_TYPE = "cuda" else: DEVICE_TYPE = "cpu" SHOW_SOURCES = True logging.info(f"Running on: {DEVICE_TYPE}") logging.info(f"Display Source Documents set to: {SHOW_SOURCES}") EMBEDDINGS = HuggingFaceInstructEmbeddings(model_name=EMBEDDING_MODEL_NAME, model_kwargs={"device": DEVICE_TYPE}) # load the vectorstore DB = Chroma( persist_directory=PERSIST_DIRECTORY, embedding_function=EMBEDDINGS, client_settings=CHROMA_SETTINGS, ) RETRIEVER = DB.as_retriever() LLM = load_model(device_type=DEVICE_TYPE, model_id=MODEL_ID, model_basename=MODEL_BASENAME) prompt, memory = get_prompt_template(promptTemplate_type="llama", history=False) template = """Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer. Use three sentences maximum. Keep the answer as concise as possible. Always say "thanks for asking!" at the end of the answer. {context} Question: {question} Helpful Answer:""" QA_CHAIN_PROMPT = PromptTemplate.from_template(template) QA = RetrievalQA.from_chain_type( llm=LLM, chain_type="stuff", retriever=RETRIEVER, return_source_documents=SHOW_SOURCES, chain_type_kwargs={ "prompt": QA_CHAIN_PROMPT, }, ) class Predict(BaseModel): prompt: str app = FastAPI(title="homepage-app") api_app = FastAPI(title="api app") app.mount("/api", api_app, name="api") app.mount("/", StaticFiles(directory="static",html = True), name="static") @api_app.post('/predict') async def predict(data: Predict): user_prompt = data.prompt if user_prompt: # print(f'User Prompt: {user_prompt}') # Get the answer from the chain res = QA(user_prompt) answer, docs = res["result"], res["source_documents"] prompt_response_dict = { "Prompt": user_prompt, "Answer": answer, } prompt_response_dict["Sources"] = [] for document in docs: prompt_response_dict["Sources"].append( (os.path.basename(str(document.metadata["source"])), str(document.page_content)) ) return {"response": prompt_response_dict} else: raise HTTPException(status_code=400, detail="Prompt Incorrect") @api_app.get("/run_ingest") def run_ingest_route(): try: if os.path.exists(PERSIST_DIRECTORY): try: shutil.rmtree(PERSIST_DIRECTORY) except OSError as e: raise HTTPException(status_code=500, detail=f"Error: {e.filename} - {e.strerror}.") else: raise HTTPException(status_code=500, detail="The directory does not exist") run_langest_commands = ["python", "ingest.py"] if DEVICE_TYPE == "cpu": run_langest_commands.append("--device_type") run_langest_commands.append(DEVICE_TYPE) result = subprocess.run(run_langest_commands, capture_output=True) if result.returncode != 0: raise HTTPException(status_code=400, detail="Script execution failed: {}") # load the vectorstore DB = Chroma( persist_directory=PERSIST_DIRECTORY, embedding_function=EMBEDDINGS, client_settings=CHROMA_SETTINGS, ) RETRIEVER = DB.as_retriever() prompt, memory = get_prompt_template(promptTemplate_type="llama", history=False) QA = RetrievalQA.from_chain_type( llm=LLM, chain_type="stuff", retriever=RETRIEVER, return_source_documents=SHOW_SOURCES, chain_type_kwargs={ "prompt": prompt, }, ) response = "Script executed successfully: {}".format(result.stdout.decode("utf-8")) return {"response": response} except Exception as e: raise HTTPException(status_code=500, detail=f"Error occurred: {str(e)}")