Update app.py

app.py

@@ -1,74 +1,20 @@
 import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM
-import spaces
-import sentencepiece
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 
-"""
-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
-"""
-try:
-    tokenizer = AutoTokenizer.from_pretrained("Yoxas/autotrain-phi3-statistical", trust_remote_code=True)
-    model = AutoModelForCausalLM.from_pretrained("Yoxas/autotrain-phi3-statistical", trust_remote_code=True)
-except Exception as e:
-    print(f"Error loading model: {e}")
-    tokenizer = None
-    model = None
+# Load the tokenizer and model
+tokenizer = AutoTokenizer.from_pretrained("Yoxas/autotrain-phi3-statistical", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained("Yoxas/autotrain-phi3-statistical", trust_remote_code=True)
 
-@spaces.GPU(duration=120)
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    if tokenizer is None or model is None:
-        yield "Error: Model not loaded properly."
-        return
+# Use a pipeline as a high-level helper
+pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, trust_remote_code=True)
 
-    messages = [{"role": "system", "content": system_message}]
+# Define the chatbot function
+def chatbot(input_text):
+    response = pipe(input_text, max_length=150, num_return_sequences=1)
+    return response[0]['generated_text']
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Create the Gradio interface
+interface = gr.Interface(fn=chatbot, inputs="text", outputs="text", title="Research Paper Abstract Chatbot")
 
-    messages.append({"role": "user", "content": message})
-
-    inputs = tokenizer(messages, return_tensors="pt", padding=True, truncation=True)
-
-    response = ""
-
-    for i in range(max_tokens):
-        try:
-            outputs = model.generate(inputs.input_ids, attention_mask=inputs.attention_mask, max_length=inputs.input_ids.shape[-1] + 1, temperature=temperature, top_p=top_p)
-            token = tokenizer.decode(outputs[0, -1:], skip_special_tokens=True)
-
-            response += token
-            yield response
-        except StopIteration:
-            break
-
-"""
-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)",
-        ),
-    ],
-)
-
-if __name__ == "__main__":
-    demo.launch()
+# Launch the Gradio app
+interface.launch()