Update app.py

app.py

@@ -1,64 +1,86 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM, TextStreamer
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Load model & tokenizer
+model_id = "sajeewa/empathy-chat-gemma"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id,
+    torch_dtype=torch.float16,
+    device_map="auto"
+)
 
+MAX_TOKENS = 2048
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# System prompt
+system_prompt = {
+    "role": "system",
+    "content": (
+        "You are an empathetic AI and your friend. Always give lovely caring messages. "
+        "Understand the user's feelings, then provide a caring response. "
+        "Talk like a sweet friend using words like 'baby', 'cutie', etc. "
+        "Use emojis when helpful. Try to continue the conversation in a gentle, emotional tone."
+    )
+}
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Initialize chat history
+chat_history = [system_prompt]
 
-    messages.append({"role": "user", "content": message})
+# Define a function to generate responses
+def respond(user_input, history):
+    global chat_history
 
-    response = ""
+    # Add user message
+    chat_history.append({"role": "user", "content": user_input})
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    # Token length control
+    chat_prompt = tokenizer.apply_chat_template(chat_history, tokenize=False)
+    while len(tokenizer(chat_prompt).input_ids) > MAX_TOKENS:
+        chat_history.pop(1)  # Remove oldest non-system message
+        chat_prompt = tokenizer.apply_chat_template(chat_history, tokenize=False)
 
-        response += token
-        yield response
+    # Prepare model input
+    inputs = tokenizer(chat_prompt, return_tensors="pt").to(model.device)
 
+    # Generate response
+    output = model.generate(
+        **inputs,
+        max_new_tokens=128,
+        temperature=0.7,
+        top_p=0.95,
+        top_k=50,
+        do_sample=True,
+    )
+    response_text = tokenizer.decode(output[0], skip_special_tokens=True)
+    new_response = response_text[len(chat_prompt):].strip()
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+    # Add assistant's response to history
+    chat_history.append({"role": "assistant", "content": new_response})
+
+    # Show full conversation
+    history.append((user_input, new_response))
+    return history, history
+
+# Define reset function
+def reset_chat():
+    global chat_history
+    chat_history = [system_prompt]
+    return [], []
+
+# Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("## 💬 Empathy Chat with Gemma")
+    chatbot = gr.Chatbot()
+    with gr.Row():
+        msg = gr.Textbox(label="Your Message", placeholder="Tell me how you feel...")
+    with gr.Row():
+        send = gr.Button("Send")
+        clear = gr.Button("Reset Chat")
 
+    send.click(fn=respond, inputs=[msg, chatbot], outputs=[chatbot, chatbot])
+    clear.click(fn=reset_chat, outputs=[chatbot, chatbot])
+    msg.submit(fn=respond, inputs=[msg, chatbot], outputs=[chatbot, chatbot])
 
-if __name__ == "__main__":
-    demo.launch()
+# Launch the app
+demo.launch()