Update app.py

app.py

@@ -2,19 +2,20 @@ import gradio as gr
 from langchain_community.document_loaders import PyPDFLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 from langchain_community.vectorstores import Chroma
-from langchain.chains import ConversationalRetrievalChain
 from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_huggingface import HuggingFacePipeline
+from langchain.chains import ConversationalRetrievalChain
 from langchain.memory import ConversationBufferMemory
 
 from pathlib import Path
 import chromadb
 from unidecode import unidecode
 
-from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
 import re
 
 # Constants
-LLM_MODEL = "facebook/bart-large-cnn"  # Changed to a model with larger response capabilities
+LLM_MODEL = "facebook/bart-large-cnn"  # Using a model with larger response capabilities
 LLM_MAX_TOKEN = 512
 DB_CHUNK_SIZE = 512
 CHUNK_OVERLAP = 24
@@ -49,7 +50,7 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, pr
     
     tokenizer = AutoTokenizer.from_pretrained(llm_model)
     model = AutoModelForSeq2SeqLM.from_pretrained(llm_model)
-    pipe = pipeline("summarization", model=model, tokenizer=tokenizer)
+    pipe = HuggingFacePipeline(pipeline("summarization", model=model, tokenizer=tokenizer))
 
     progress(0.75, desc="Defining buffer memory...")
     memory = ConversationBufferMemory(