Update app.py

app.py

@@ -36,11 +36,24 @@ with st.sidebar:
 if "messages" not in st.session_state:
     st.session_state.messages = []
 
+# File processing function
 @st.cache_data
 def process_file(uploaded_file):
-    # Existing file processing logic
-    pass
+    if uploaded_file is None:
+        return ""
+    
+    try:
+        if uploaded_file.type == "application/pdf":
+            pdf_reader = PyPDF2.PdfReader(uploaded_file)
+            return "\n".join([page.extract_text() for page in pdf_reader.pages])
+        elif uploaded_file.type == "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet":
+            df = pd.read_excel(uploaded_file)
+            return df.to_markdown()
+    except Exception as e:
+        st.error(f"📄 Error processing file: {str(e)}")
+        return ""
 
+# Model loading function
 @st.cache_resource
 def load_model(hf_token):
     try:
@@ -48,16 +61,13 @@ def load_model(hf_token):
             st.error("🔐 Authentication required! Please provide a Hugging Face token.")
             return None
         
-        # Login to Hugging Face Hub
         login(token=hf_token)
         
-        # Load tokenizer
         tokenizer = AutoTokenizer.from_pretrained(
             MODEL_NAME,
             token=hf_token
         )
         
-        # Load model with KV caching support
         model = AutoModelForCausalLM.from_pretrained(
             MODEL_NAME,
             device_map="auto",
@@ -71,7 +81,43 @@ def load_model(hf_token):
         st.error(f"🤖 Model loading failed: {str(e)}")
         return None
 
-# In the main chat handling section:
+# Generation function with KV caching
+def generate_with_kv_cache(prompt, file_context, use_cache=True):
+    full_prompt = f"Analyze this context:\n{file_context}\n\nQuestion: {prompt}\nAnswer:"
+    
+    streamer = TextIteratorStreamer(
+        tokenizer, 
+        skip_prompt=True, 
+        skip_special_tokens=True
+    )
+    
+    inputs = tokenizer(full_prompt, return_tensors="pt").to(model.device)
+    
+    generation_kwargs = {
+        **inputs,
+        "max_new_tokens": 1024,
+        "temperature": 0.7,
+        "top_p": 0.9,
+        "repetition_penalty": 1.1,
+        "do_sample": True,
+        "use_cache": use_cache,
+        "streamer": streamer
+    }
+    
+    Thread(target=model.generate, kwargs=generation_kwargs).start()
+    return streamer
+
+# Display chat messages
+for message in st.session_state.messages:
+    try:
+        avatar = "👤" if message["role"] == "user" else "🤖"
+        with st.chat_message(message["role"], avatar=avatar):
+            st.markdown(message["content"])
+    except:
+        with st.chat_message(message["role"]):
+            st.markdown(message["content"])
+
+# Chat input handling
 if prompt := st.chat_input("Ask your inspection question..."):
     if not hf_token:
         st.error("🔑 Authentication required!")
@@ -89,7 +135,6 @@ if prompt := st.chat_input("Ask your inspection question..."):
     model = st.session_state.model
     tokenizer = st.session_state.tokenizer
     
-    
     # Add user message
     with st.chat_message("user", avatar="👤"):
         st.markdown(prompt)