OpenNMT

Tqdm with translate_batch()

Hello, Colleagues!

Any ideas how to use tqdm with translate_batch() for large inputs? The only solution I could think of is to split large inputs into chunks; however, I do not think it is perfect.

Thanks!
Yasmin

Hi,

I don’t think there is another solution to use tqdm. You should split your input before calling translate_batch.

We can’t use tqdm inside translate_batch since the underlying implementation is in C++.

1 Like