Update app.py

app.py

@@ -10,7 +10,8 @@ def generate_text(Text,Max_length,Temperature):
   tokenizer.pad_token_id=tokenizer.eos_token_id
   with torch.no_grad():
     input_ids = tokenizer(Text, return_tensors="pt")["input_ids"].to('cpu')
-    output=model.generate(input_ids,max_new_tokens=Max_length,do_sample=True, temperature=Temperature, top_p=0.90,top_k=10)
+    attn_mask=tokenizer(Text, return_tensors="pt")["attention_mask"].to('cpu')
+    output=model.generate(input_ids=input_ids,attention_mask=attn_mask,max_new_tokens=Max_length,do_sample=True, temperature=Temperature, top_p=0.90,top_k=10)
     return tokenizer.decode(output[0])
 demo=gr.Interface(
     generate_text,