Update app.py

app.py

@@ -10,7 +10,9 @@ import math
 from transformers import pipeline
 from langchain.prompts import ChatPromptTemplate
 from langchain_community.llms import HuggingFaceHub
+from langchain.chains.summarize import load_summarize_chain
 import re
+from dotenv import load_dotenv
 # import json
 
 # st.config(PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION="python")
@@ -95,13 +97,36 @@ def prompt_engineer(text, longtext, query):
     BULLET POINT SUMMARY:
     """
     # Load the summarization pipeline with the specified model
-    summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
+    # summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
 
     # Generate the prompt
-    prompt = summary_prompt_template.format(text=text)
+    # prompt = summary_prompt_template.format(text=text)
 
     # Generate the summary
-    summary = summarizer(prompt, max_length=1024, min_length=50)[0]["summary_text"]
+    # summary = summarizer(prompt, max_length=1024, min_length=50)[0]["summary_text"]
+
+    try:
+        llm = HuggingFaceHub(
+            repo_id="meta-llama/Meta-Llama-3-8B-Instruct", model_kwargs={"temperature": 0, "max_new_tokens": 256, "task":"text-generation"}
+        )
+        st.write("llm connection started..")
+    except Exception as e:
+        st.error(f"Error invoke: {e}")
+
+    from langchain.chains.combine_documents import create_stuff_documents_chain
+    from langchain.chains.llm import LLMChain
+    from langchain_core.prompts import ChatPromptTemplate
+    
+    # Define prompt
+    prompt = ChatPromptTemplate.from_messages(
+        [("system", summary_prompt_template)]
+    )
+    
+    # Instantiate chain
+    chain = create_stuff_documents_chain(llm, prompt)
+    
+    # Invoke chain
+    summary = chain.invoke({"text": longtext})
     
     with st.sidebar:
         st.divider()
@@ -130,9 +155,9 @@ def prompt_engineer(text, longtext, query):
     result = ""
     
     try:
-        llm = HuggingFaceHub(
-            repo_id="meta-llama/Meta-Llama-3-8B-Instruct", model_kwargs={"temperature": 0.1, "max_new_tokens": 256, "task":"text-generation"}
-        )
+        # llm = HuggingFaceHub(
+        #     repo_id="meta-llama/Meta-Llama-3-8B-Instruct", model_kwargs={"temperature": 0, "max_new_tokens": 256, "task":"text-generation"}
+        # )
         response_text = llm.invoke(prompt)
         escaped_query = re.escape(query)
         result = re.split(f'Answer the question based on the above context: {escaped_query}\n',response_text)[-1]