Update app.py

app.py

@@ -1,28 +1,30 @@
-from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
-
-tokenizer = AutoTokenizer.from_pretrained("google/flan-t5-large")
+import os
+import gradio as gr
+import torch
+import numpy as np
+from transformers import pipeline
 
-model = AutoModelForSeq2SeqLM.from_pretrained("google/flan-t5-large", device_map="auto",  cache_dir="cache", offload_folder="offload" )
+import torch
 
-def generate(input_text):
-  input_ids = tokenizer(input_text, return_tensors="pt")
-  output = model.generate(input_ids, max_length=70)
-  return tokenizer.decode(output[0], skip_special_tokens=True)
+pipe_flan = pipeline("text2text-generation", model="google/flan-t5-large", model_kwargs={"device_map": "auto"})
 
-    #@title GUI
-import gradio as gr
+examples = []
 
-title = "Flan T5 :)"
+title = "Flan T5 and Vanilla T5"
+description = ""
 
 def inference(text):
-  return generate(text)
+  output_flan = pipe_flan(text, max_length=100)[0]["generated_text"]
+  return [output_flan]
 
 io = gr.Interface(
   inference,
   gr.Textbox(lines=3),
   outputs=[
-    gr.Textbox(lines=3, label="Flan T5")
+    gr.Textbox(lines=3, label="Flan T5"),
   ],
   title=title,
+  description=description,
+  examples=examples
 )
-io.launch(share=False,debug=False)
+io.launch()