Update app.py

app.py

@@ -1,89 +1,148 @@
 import gradio as gr
-# from langchain.llms import GooglePalm
-from langchain_google_genai import GoogleGenerativeAI
-from langchain.document_loaders.csv_loader import CSVLoader
-from langchain_huggingface import HuggingFaceEmbeddings
-from langchain.vectorstores import FAISS
+from langchain_community.document_loaders import CSVLoader  # Changed import
+from langchain_community.vectorstores import FAISS  # Changed import
 from langchain.prompts import PromptTemplate
 from langchain.chains import RetrievalQA
+from langchain.llms import HuggingFaceLLM  # Adjusted for correct instantiation
 import warnings
 from huggingface_hub import login
 import os
-
-
 from transformers import pipeline
-llm = pipeline("feature-extraction", model="mixedbread-ai/mxbai-embed-large-v1") 
-
-# from transformers import AutoModel
-# llm = AutoModel.from_pretrained("Alibaba-NLP/gte-large-en-v1.5", trust_remote_code=True) 
-
-# LLAMA
-# from transformers import AutoModelForCausalLM, AutoTokenizer
-# from transformers import pipeline
-
-# hf_token = os.environ['llama_token']
-
-# login(token=hf_token)
-
-# llm = pipeline("text-generation", model="meta-llama/Meta-Llama-3-8B-Instruct")
-# llm = pipeline("text-generation", model = "meta-llama/Meta-Llama-3-70B-Instruct")
-
-# MISTRAL
-# llm = pipeline("text-generation", model="mistralai/Mixtral-8x22B-Instruct-v0.1")
 
+# Initialize the LLM using pipeline
+llm = pipeline("text-generation", model="meta-llama/Meta-Llama-3-8B-Instruct")  # Adjusted initialization
 
-# TO USE GOOGLE MODELS
-# api_key = "AIzaSyCdM_aAIsW_nPbjarOF83mbX1_z1cVX2_M"
-
-# llm = GoogleGenerativeAI(model="models/text-bison-001", google_api_key=api_key)
-# llm = GooglePalm(google_api_key = api_key, temperature=0.7)
-
-# LOADING CSV FILE
-loader = CSVLoader(file_path='aiotsmartlabs_faq.csv', source_column = 'prompt')
+# Load CSV file
+loader = CSVLoader(file_path='aiotsmartlabs_faq.csv', source_column='prompt')
 data = loader.load()
 
-# SUPPRESSING WARNINGS
+# Suppress warnings
 warnings.filterwarnings("ignore", category=UserWarning, message="TypedStorage is deprecated")
 warnings.filterwarnings("ignore", category=FutureWarning, message="`resume_download` is deprecated")
 
-
-# EMBEDDING MODEL
+# Embedding model
 model_name = "BAAI/bge-m3"
-instructor_embeddings = HuggingFaceEmbeddings(model_name=model_name)
+instructor_embeddings = HuggingFaceLLM(model_name=model_name)  # Adjusted for correct instantiation
 
 # Create FAISS vector store from documents
 vectordb = FAISS.from_documents(documents=data, embedding=instructor_embeddings)
 retriever = vectordb.as_retriever()
 
+# Define the prompt template
 prompt_template = """Given the following context and a question, generate an answer based on the context only.
-
 In the answer try to provide as much text as possible from "response" section in the source document context without making much changes.
 If somebody asks "Who are you?" or a similar phrase, state "I am Rishi's assistant built using a Large Language Model!"
 If the answer is not found in the context, kindly state "I don't know. Please ask Rishi on Discord. Discord Invite Link: https://discord.gg/6ezpZGeCcM. Or email at [email protected]" Don't try to make up an answer.
-
 CONTEXT: {context}
-
 QUESTION: {question}"""
 
 PROMPT = PromptTemplate(
-    template = prompt_template, input_variables = ["context", "question"]
+    template=prompt_template, input_variables=["context", "question"]
 )
 
+# Initialize the RetrievalQA chain
+chain = RetrievalQA.from_chain_type(llm=llm,  # Adjusted initialization
+                                    chain_type="stuff",
+                                    retriever=retriever,
+                                    input_key="query",
+                                    return_source_documents=True,
+                                    chain_type_kwargs={"prompt": PROMPT})
 
-chain = RetrievalQA.from_chain_type(llm = llm,
-            chain_type="stuff",
-            retriever=retriever,
-            input_key="query",
-            return_source_documents=True,
-            chain_type_kwargs = {"prompt": PROMPT})
-
+# Define the chat response function
 def chatresponse(message, history):
     output = chain(message)
     return output['result']
 
+# Launch the Gradio chat interface
 gr.ChatInterface(chatresponse).launch()
 
 
+# import gradio as gr
+# # from langchain.llms import GooglePalm
+# from langchain_google_genai import GoogleGenerativeAI
+# from langchain.document_loaders.csv_loader import CSVLoader
+# from langchain_huggingface import HuggingFaceEmbeddings
+# from langchain.vectorstores import FAISS
+# from langchain.prompts import PromptTemplate
+# from langchain.chains import RetrievalQA
+# import warnings
+# from huggingface_hub import login
+# import os
+
+
+# from transformers import pipeline
+# llm = pipeline("feature-extraction", model="mixedbread-ai/mxbai-embed-large-v1") 
+
+# # from transformers import AutoModel
+# # llm = AutoModel.from_pretrained("Alibaba-NLP/gte-large-en-v1.5", trust_remote_code=True) 
+
+# # LLAMA
+# # from transformers import AutoModelForCausalLM, AutoTokenizer
+# # from transformers import pipeline
+
+# # hf_token = os.environ['llama_token']
+
+# # login(token=hf_token)
+
+# # llm = pipeline("text-generation", model="meta-llama/Meta-Llama-3-8B-Instruct")
+# # llm = pipeline("text-generation", model = "meta-llama/Meta-Llama-3-70B-Instruct")
+
+# # MISTRAL
+# # llm = pipeline("text-generation", model="mistralai/Mixtral-8x22B-Instruct-v0.1")
+
+
+# # TO USE GOOGLE MODELS
+# # api_key = "AIzaSyCdM_aAIsW_nPbjarOF83mbX1_z1cVX2_M"
+
+# # llm = GoogleGenerativeAI(model="models/text-bison-001", google_api_key=api_key)
+# # llm = GooglePalm(google_api_key = api_key, temperature=0.7)
+
+# # LOADING CSV FILE
+# loader = CSVLoader(file_path='aiotsmartlabs_faq.csv', source_column = 'prompt')
+# data = loader.load()
+
+# # SUPPRESSING WARNINGS
+# warnings.filterwarnings("ignore", category=UserWarning, message="TypedStorage is deprecated")
+# warnings.filterwarnings("ignore", category=FutureWarning, message="`resume_download` is deprecated")
+
+
+# # EMBEDDING MODEL
+# model_name = "BAAI/bge-m3"
+# instructor_embeddings = HuggingFaceEmbeddings(model_name=model_name)
+
+# # Create FAISS vector store from documents
+# vectordb = FAISS.from_documents(documents=data, embedding=instructor_embeddings)
+# retriever = vectordb.as_retriever()
+
+# prompt_template = """Given the following context and a question, generate an answer based on the context only.
+
+# In the answer try to provide as much text as possible from "response" section in the source document context without making much changes.
+# If somebody asks "Who are you?" or a similar phrase, state "I am Rishi's assistant built using a Large Language Model!"
+# If the answer is not found in the context, kindly state "I don't know. Please ask Rishi on Discord. Discord Invite Link: https://discord.gg/6ezpZGeCcM. Or email at [email protected]" Don't try to make up an answer.
+
+# CONTEXT: {context}
+
+# QUESTION: {question}"""
+
+# PROMPT = PromptTemplate(
+#     template = prompt_template, input_variables = ["context", "question"]
+# )
+
+
+# chain = RetrievalQA.from_chain_type(llm = llm,
+#             chain_type="stuff",
+#             retriever=retriever,
+#             input_key="query",
+#             return_source_documents=True,
+#             chain_type_kwargs = {"prompt": PROMPT})
+
+# def chatresponse(message, history):
+#     output = chain(message)
+#     return output['result']
+
+# gr.ChatInterface(chatresponse).launch()
+
+
 
 # import gradio as gr