Update app.py

app.py

@@ -4,17 +4,19 @@ import torch
 import os
 from dotenv import load_dotenv
 
+# Load environment variables
 load_dotenv()
-
 api_key = os.getenv("api_key")
+
 # App title and description
 st.title("I am Your GrowBuddy 🌱")
 st.write("Let me help you start gardening. Let's grow together!")
 
+# Function to load model
 def load_model():
     try:
-        tokenizer = AutoTokenizer.from_pretrained("KhunPop/Gardening",  use_auth_token=api_key)
-        model = AutoModelForCausalLM.from_pretrained("google/gemma-2b-it",  use_auth_token=api_key)
+        tokenizer = AutoTokenizer.from_pretrained("KhunPop/Gardening", use_auth_token=api_key)
+        model = AutoModelForCausalLM.from_pretrained("QuantFactory/leniachat-gemma-2b-v0-GGUF", use_auth_token=api_key)
         return tokenizer, model
     except Exception as e:
         st.error(f"Failed to load model: {e}")
@@ -30,46 +32,45 @@ if not tokenizer or not model:
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 model = model.to(device)
 
-# Initialize session state messages if not already initialized
+# Initialize session state messages
 if "messages" not in st.session_state:
     st.session_state.messages = [
         {"role": "assistant", "content": "Hello there! How can I help you with gardening today?"}
     ]
 
-# Display the conversation history
+# Display conversation history
 for message in st.session_state.messages:
     with st.chat_message(message["role"]):
         st.write(message["content"])
 
+# Function to generate response
 def generate_response(prompt):
     try:
-        # Tokenize the input prompt
+        # Tokenize input prompt with dynamic padding and truncation
         inputs = tokenizer(prompt, return_tensors="pt", truncation=True, padding=True, max_length=512).to(device)
 
-        # Ensure the model is generating properly (without a target)
-        outputs = model.generate(inputs["input_ids"], max_new_tokens=150, temperature=0.7, do_sample=True)
+        # Generate output from model
+        outputs = model.generate(inputs["input_ids"], max_new_tokens=100, temperature=0.7, do_sample=True)
 
-        # Decode the output to text
+        # Decode and return response
         response = tokenizer.decode(outputs[0], skip_special_tokens=True)
         return response
     except Exception as e:
         st.error(f"Error during text generation: {e}")
         return "Sorry, I couldn't process your request."
 
-# User input field for asking questions
+# User input field for gardening questions
 user_input = st.chat_input("Type your gardening question here:")
 
 if user_input:
-    # Display user message
     with st.chat_message("user"):
         st.write(user_input)
 
-    # Generate and display assistant's response
     with st.chat_message("assistant"):
-        with st.spinner("I'm gonna tell you..."):
+        with st.spinner("Generating your answer..."):
             response = generate_response(user_input)
             st.write(response)
 
-    # Update session state with the new conversation
+    # Update session state
     st.session_state.messages.append({"role": "user", "content": user_input})
     st.session_state.messages.append({"role": "assistant", "content": response})