Update app.py

app.py

@@ -61,25 +61,20 @@ def translate(
     print(f'Text is - {source_text}')
     
     prompt = Prompt_template(source_text, source_lang, target_lang)
-    input_ids = tokenizer(prompt, return_tensors="pt").to(model.device)
+    input_ids = tokenizer(prompt, return_tensors="pt").input_ids
         
-    streamer = TextIteratorStreamer(tokenizer, **{"skip_special_tokens": True, "skip_prompt": True, 'clean_up_tokenization_spaces':False,})
-
     generate_kwargs = dict(
         input_ids=input_ids,
-        streamer=streamer,
         max_length=max_length, 
         do_sample=True, 
         temperature=temperature,
     )
-    
-    thread = Thread(target=model.generate, kwargs=generate_kwargs)
-    thread.start()
 
-    buffer = ""
-    for new_text in streamer:
-        buffer += new_text
-        yield buffer
+    generate_ids = model.generate(**generate_kwargs)
+    
+    resp = tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+    
+    return resp
 
 CSS = """
     h1 {