advanced the gpt model to faster inference

app.py

@@ -1,9 +1,13 @@
-from transformers import pipeline
+from transformers import pipeline, AutoTokenizer
 import gradio as gr
 
 model_id = "gpt2-large"
 
-pipe = pipeline("text-generation", model=model_id)
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id)
+model.to_bettertransformer()
+
+pipe = pipeline("text-generation", model=model, tokenizer= tokenizer)
 
 def gpt(prompt, top_k, penalty_alpha):
     return pipe(prompt, top_k=top_k, penalty_alpha=penalty_alpha)[0]["generated_text"]

requirements.txt

@@ -1,3 +1,4 @@
 transformers
 gradio
 torch
+optimum