Update app.py

app.py

@@ -10,15 +10,25 @@ from langchain.vectorstores import FAISS
 from langchain.chains.question_answering import load_qa_chain
 from langchain.prompts import PromptTemplate
 import whisper
-from groq import GroqAPI
+import requests
 from dotenv import load_dotenv
 
-# Step 2: Load environment variable
+# Step 2: Load environment variables
 load_dotenv()
 groq_api_key = os.getenv("GROQ_API_KEY")
 
-# Step 3: Initialize Groq API
-groq = GroqAPI(api_key=groq_api_key)
+# Step 3: Custom function to interact with the Groq API
+def get_groq_embeddings(text_chunks):
+    url = "https://api.groq.com/your-endpoint"  # Replace with the correct Groq API endpoint
+    headers = {"Authorization": f"Bearer {groq_api_key}"}
+    payload = {"text_chunks": text_chunks}
+    response = requests.post(url, json=payload, headers=headers)
+    
+    if response.status_code == 200:
+        return response.json()["embeddings"]
+    else:
+        st.error(f"Error: {response.status_code} - {response.text}")
+        return None
 
 # Step 4: Function to read files and extract text
 def extract_text(file):
@@ -59,15 +69,18 @@ def get_text_chunks(text):
 
 # Step 6: Function for converting chunks into embeddings and saving the FAISS index
 def get_vector_store(text_chunks):
-    embeddings = groq.get_embeddings(text_chunks)
-    vector_store = FAISS.from_texts(text_chunks, embedding=embeddings)
-    
-    # Ensure the directory exists
-    if not os.path.exists("faiss_index"):
-        os.makedirs("faiss_index")
+    embeddings = get_groq_embeddings(text_chunks)
+    if embeddings:
+        vector_store = FAISS.from_texts(text_chunks, embedding=embeddings)
         
-    vector_store.save_local("faiss_index")
-    print("FAISS index saved successfully.")
+        # Ensure the directory exists
+        if not os.path.exists("faiss_index"):
+            os.makedirs("faiss_index")
+            
+        vector_store.save_local("faiss_index")
+        print("FAISS index saved successfully.")
+    else:
+        st.error("Failed to retrieve embeddings from Groq API.")
 
 # Step 7: Function to implement the Groq Model
 def get_conversational_chain():
@@ -76,22 +89,26 @@ def get_conversational_chain():
     the provided context, just say, "The answer is not available in the context." Do not provide a wrong answer.\n\n
     Context:\n {context}\n
     Question: \n{question}\n
-
     Answer:
     """
-    model = groq.get_chat_model("llama3-8b-8192")  # Replace with your Groq model ID
+    # Assuming we use the Groq API for the model as well
+    # Replace with your Groq model call or other LLM API
+    model = get_groq_embeddings  # Placeholder for the actual model call
     prompt = PromptTemplate(template=prompt_template, input_variables=["context", "question"])
     chain = load_qa_chain(model, chain_type="stuff", prompt=prompt)
     return chain
 
 # Step 8: Function to take inputs from user and generate response
 def user_input(user_question):
-    embeddings = groq.get_embeddings([user_question])
-    new_db = FAISS.load_local("faiss_index", embeddings, allow_dangerous_deserialization=True)
-    docs = new_db.similarity_search(user_question)
-    chain = get_conversational_chain()
-    response = chain({"input_documents": docs, "question": user_question}, return_only_outputs=True)
-    return response["output_text"]
+    embeddings = get_groq_embeddings([user_question])
+    if embeddings:
+        new_db = FAISS.load_local("faiss_index", embeddings, allow_dangerous_deserialization=True)
+        docs = new_db.similarity_search(user_question)
+        chain = get_conversational_chain()
+        response = chain({"input_documents": docs, "question": user_question}, return_only_outputs=True)
+        return response["output_text"]
+    else:
+        return "Failed to retrieve response from Groq API."
 
 # Step 9: Streamlit App
 def main():