Update app.py

app.py

@@ -1,7 +1,11 @@
 import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 
-model_path = "Wonder-Griffin/ShorseyBeerLeague"
+# Use a pipeline as a high-level helper for text generation
+pipe = pipeline("text-generation", model="Wonder-Griffin/ShorseyBeerLeague")
+
+# Assuming `model_path` is the Hugging Face model hub path or a local directory
+model_path = "Wonder-Griffin/ShorseyBeerLeague"  # Define this as needed
 
 tokenizer = AutoTokenizer.from_pretrained(model_path)
 model = AutoModelForCausalLM.from_pretrained(
@@ -18,6 +22,7 @@ def respond(
     temperature,
     top_p,
 ):
+    # Building the conversation history for the model
     messages = [{"role": "system", "content": system_message}]
 
     for val in history:
@@ -27,16 +32,26 @@ def respond(
             messages.append({"role": "assistant", "content": val[1]})
 
     messages.append({"role": "user", "content": message})
+    
+    # Tokenize the input message
+    input_text = " ".join([msg["content"] for msg in messages if msg["role"] == "user"])
+    input_ids = tokenizer.encode(input_text, return_tensors="pt")
+    
+    # Generate a response from the model
+    output_ids = model.generate(
+        input_ids.to("cuda"),
+        max_new_tokens=max_tokens,
+        temperature=temperature,
+        top_p=top_p,
+        do_sample=True
+    )
+    
+    # Decode the generated tokens into a response
+    response = tokenizer.decode(output_ids[0], skip_special_tokens=True)
 
-    input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
-    output_ids = model.generate(input_ids.to('cuda'))
-    respond = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
-
-    return respond
+    return response
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Gradio interface setup
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -51,4 +66,7 @@ demo = gr.ChatInterface(
             label="Top-p (nucleus sampling)",
         ),
     ],
-)
+)
+
+if __name__ == "__main__":
+    demo.launch()