Update app.py

app.py

@@ -5,6 +5,7 @@ from langchain_community.document_loaders import DirectoryLoader, TextLoader
 from langchain_community.vectorstores import FAISS
 from langchain_community.embeddings import HuggingFaceEmbeddings
 from langchain.text_splitter import RecursiveCharacterTextSplitter
+from transformers import AutoTokenizer
 
 # Step 1: Extract ZIP
 def extract_zip(zip_path, extract_to):
@@ -18,11 +19,28 @@ def extract_zip(zip_path, extract_to):
 def load_knowledge_base(folder_path):
     loader = DirectoryLoader(folder_path, glob="*.md", loader_cls=TextLoader)
     docs = loader.load()
+    if not docs:
+        st.error("❌ No documents found in the knowledge base folder.")
+        return None
+
     splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
     split_docs = splitter.split_documents(docs)
-    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/paraphrase-MiniLM-L6-v2")
-    db = FAISS.from_documents(split_docs, embeddings)
-    return db
+    if not split_docs:
+        st.error("❌ Failed to split documents.")
+        return None
+
+    # Use AutoTokenizer to load the appropriate tokenizer for the model
+    model_name = "sentence-transformers/paraphrase-MiniLM-L6-v2"
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+    embeddings = HuggingFaceEmbeddings(model_name=model_name, tokenizer=tokenizer)
+
+    try:
+        db = FAISS.from_documents(split_docs, embeddings)
+        return db
+    except Exception as e:
+        st.error(f"❌ Error creating FAISS index: {e}")
+        return None
 
 # Streamlit UI
 st.title("📘 Fitlytic Chatbot")
@@ -40,10 +58,13 @@ else:
     st.stop()
 
 # Step 5: User interaction
-query = st.text_input("Ask me anything about Fitlytic:")
-if query:
-    results = db.similarity_search(query, k=1)
-    if results:
-        st.success(results[0].page_content)
-    else:
-        st.error("😕 Sorry, I couldn't find an answer. Try rephrasing it.")
+if db:
+    query = st.text_input("Ask me anything about Fitlytic:")
+    if query:
+        results = db.similarity_search(query, k=1)
+        if results:
+            st.success(results[0].page_content)
+        else:
+            st.error("😕 Sorry, I couldn't find an answer. Try rephrasing it.")
+else:
+    st.stop()