Update app.py

app.py

@@ -216,15 +216,18 @@ def unbabel(model_name, sl, tl, input_text):
     messages = [{"role": "user",
                  "content": f"Translate the following text from {sl} into {tl}.\n{sl}: {input_text}.\n{tl}:"}]
     prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=False)
-    max_new_tokens = int(len(input_text) // 3) + int(0.25 * len(input_text))
-    print(len(input_text), max_new_tokens)
+    tokenized_input = pipe.tokenizer(input_text, return_tensors="pt")
+    num_input_tokens = len(tokenized_input["input_ids"][0])
+    max_new_tokens = int(num_input_tokens) + int(0.25 * len(num_input_tokens))
+    print(len(input_text), tokenized_input, num_input_tokens, max_new_tokens)
     outputs = pipe(prompt, max_new_tokens=max_new_tokens, do_sample=False)
     translated_text = outputs[0]["generated_text"]
     markers = ["<end_of_turn>", "<|im_end|>", "<|im_start|>assistant"] # , "\n" 
     for marker in markers:
         if marker in translated_text:
             translated_text = translated_text.split(marker)[1].strip()
-        translated_text = translated_text.replace('Answer:', '', 1).strip() if translated_text.startswith('Answer:') else translated_text 
+    translated_text = translated_text.replace('Answer:', '', 1).strip() if translated_text.startswith('Answer:') else translated_text
+    translated_text = translated_text.split("Translated text:")[0].strip() if "Translated text:" in translated_text else translated_text
     return translated_text
     
 def mbart_many_to_many(model_name, sl, tl, input_text):