dyyzhmm comments

Repositories
Issues
Comments

Results 5 comments of


                                            dyyzhmm

I can't run the train.py with CASME2 dataset.

@ZYDi I make it finally. The reason is that I make a mistake with the dataset CASME2 and CASME2_OPtical_Flow.

I can't run the train.py with CASME2 dataset.

@ZYDi [email protected] That's my mailbox.

The size of per_device_train_batch_size

same question +1

faster_whisper batch operation time-consuming

> Can you move the input features to the model device before measuring the execution time ? 10 samples. Before changes ``` def get_ctranslate2_storage(segment: np.ndarray) -> ctranslate2.StorageView: segment = np.ascontiguousarray(segment)...

faster_whisper batch operation time-consuming

@funboarder13920 hi, are you using `torch.profiler` or `python cProfile` to obtain the analysis results mentioned above?