Yuhui Zhang comments

Results 9 comments of


                                            Yuhui Zhang

Pretokenized and MWT

Hi @501Good, thank you for your interest in contributing Stanza! The solution and code generally look good to me. Can you list 2-3 test cases you used (best if runnable...

Leverage multiple gpus for dependency parse model?

Hi @masonedmison, supporting multiple GPUs and maximizing processing speed is our future direction, but unfortunately, this improvement is unlikely to release in a short period. If you are going to...

Bias not used

Thank you for pointing out! This is indeed a bug that we confirmed. However, we suppose this small bug might have little influence on model performance. We will figure it...

Low performance in many-cores systems

Hi @StarTessar, thanks for reporting this! This may due to the nature of PyTorch parallelization, as you say the problem is solved by explicitly declaring `OMP_NUM_THREADS` in the program. I...

Russian text is not broken down into sentences

Thank you for your information! The problem can be reproduced. We will try to figure it out soon.

SimCLR can not converge on CIFAR10.

I encountered the same problem when trying to use `syncbn` without `apex` support. Just removing `--syncbn` solves the problem.

Could you verify the implementation? Acc = 11% after domain adaptation

See #29

How to use different decoder model?

You can replace the GPT2 decoder code with LLaMA3 code

Please add InternVL2.5

Thank you for sharing—this model looks very strong! Would you be open to evaluating your model on our benchmark and sending us the JSON predictions? We’d love to integrate the...