File size: 1,444 Bytes
ee0f24f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
import os
from dotenv import load_dotenv
from prompts import qa_template_V0, qa_template_V1, qa_template_V2
# Load environment variables from .env file
load_dotenv()
from langchain.chat_models import ChatAnyscale
ANYSCALE_ENDPOINT_TOKEN=os.environ.get("ANYSCALE_ENDPOINT_TOKEN")
anyscale_api_key =ANYSCALE_ENDPOINT_TOKEN
llm=ChatAnyscale(anyscale_api_key=anyscale_api_key,temperature=0, model_name='mistralai/Mistral-7B-Instruct-v0.1', streaming=False)
## Create embeddings and splitter
from langchain.embeddings import HuggingFaceBgeEmbeddings
from langchain.text_splitter import RecursiveCharacterTextSplitter
# Create Embeddings
model_name = "BAAI/bge-large-en"
embedding = HuggingFaceBgeEmbeddings(
model_name = model_name,
# model_kwargs = {'device':'cuda'},
encode_kwargs = {'normalize_embeddings': True}
)
# Create Splitter
splitter = RecursiveCharacterTextSplitter(
chunk_size=1000,
chunk_overlap=100,
)
from langchain_community.vectorstores import FAISS
# persits_directory="./faiss_V04_C500_BGE_large_web_doc_with_split-final"
persits_directory="./faiss_V06_C500_BGE_large-Final"
vectorstore= FAISS.load_local(persits_directory, embedding)
# Define a custom prompt for Unser manual
from langchain.prompts import PromptTemplate
QA_PROMPT = PromptTemplate(input_variables=["context", "question"],template=qa_template_V2,)
|