Update app.py

app.py

@@ -17,7 +17,7 @@ from langchain_community.llms import HuggingFaceEndpoint
 import torch
 api_token = os.getenv("HF_TOKEN")
 
-list_llm = ["meta-llama/Meta-Llama-3-8B-Instruct", "mistralai/Mistral-7B-Instruct-v0.3"]
+list_llm = ["meta-llama/Meta-Llama-3-8B-Instruct", "mistralai/Mistral-7B-Instruct-v0.3"]  
 list_llm_simple = [os.path.basename(llm) for llm in list_llm]
 
 # Load PDF document and create doc splits
@@ -64,19 +64,32 @@ def create_db(splits, collection_name, db_type):
     return vectordb
 
 # Initialize langchain LLM chain
-def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, initial_prompt, progress=gr.Progress()):
+# Initialize langchain LLM chain
+def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, progress=gr.Progress()):
     progress(0.1, desc="Initializing HF tokenizer...")
     
     progress(0.5, desc="Initializing HF Hub...")
 
-    llm = HuggingFaceEndpoint(
-        repo_id=llm_model,
-        huggingfacehub_api_token=api_token,
-        temperature=temperature,
-        max_new_tokens=max_tokens,
-        top_k=top_k,
-    )
-
+    
+    if llm_model == "meta-llama/Meta-Llama-3-8B-Instruct":
+        llm = HuggingFaceEndpoint(
+            repo_id=llm_model,
+            huggingfacehub_api_token=api_token,
+            temperature=temperature,
+            max_new_tokens=max_tokens,
+            top_k=top_k,
+        )
+   
+    else:
+        
+        llm = HuggingFaceEndpoint(
+            repo_id=llm_model,
+            huggingfacehub_api_token=api_token,
+            temperature=temperature,
+            max_new_tokens=max_tokens,
+            top_k=top_k,        
+        )
+    
     progress(0.75, desc="Defining buffer memory...")
     memory = ConversationBufferMemory(
         memory_key="chat_history",
@@ -88,12 +101,11 @@ def initialize_llmchain(llm_model, temperature, max_tokens, top_k, vector_db, in
     qa_chain = ConversationalRetrievalChain.from_llm(
         llm,
         retriever=retriever,
-        chain_type="stuff",
+        chain_type="stuff", 
         memory=memory,
         return_source_documents=True,
         verbose=False,
     )
-    qa_chain({"question": initial_prompt})  # Initialize with the initial prompt
     progress(0.9, desc="Done!")
     return qa_chain