Update app.py

app.py

@@ -1,9 +1,7 @@
 import gradio as gr
 import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer, StoppingCriteria, StoppingCriteriaList, TextIteratorStreamer, GPTQConfig, TrainingArguments
-from threading import Thread
+from transformers import AutoTokenizer, GenerationConfig
 from peft import AutoPeftModelForCausalLM
-from transformers import GenerationConfig
 tokenizer = AutoTokenizer.from_pretrained("izh97/zephyr-beta-climate-change-assistant")
 model = AutoPeftModelForCausalLM.from_pretrained(
     "izh97/zephyr-beta-climate-change-assistant",
@@ -35,4 +33,6 @@ def ask(text):
                            return_dict_in_generate=True)
 
   tokens = outputs.sequences[0, input_length:]
-  return tokenizer.decode(tokens)
+  return tokenizer.decode(tokens)
+iface = gr.Interface(fn=ask, inputs="text", outputs="text")
+iface.launch()