How to get the ${nshard} value and ${rank} value? Are they random numbers?

[SE-Nickjackson]

Hi authors, thank you for the great work! When I preprocess the dataset, I don't know what values ${nshard} and ${rank} should take. Can you help me out？

[chevalierNoir]

Hi,

${nshard} splits the whole data into ${nshard} parts and ${rank} is the job id processing the ${rank}th part. For example.,python lrs3_prepare.py --lrs3 ${lrs3} --ffmpeg /path/to/ffmpeg --rank ${rank} --nshard ${nshard} --step ${step} is the job for processing the ${rank}-th shard of the LRS3 dataset. To process the whole dataset, you need to launch ${nshard} jobs with ${rank} varying from 0 to ${nshard}-1.

[SE-Nickjackson]

Ok, thanks

[shakeel608]

Hi after running

python lrs3_prepare.py --lrs3 lrs3/ --ffmpeg /path/to/ffmpeg --rank 10 --nshard 20 --step 1 I am getting

if word_intervals[-1][-1] < max_duration: IndexError: list index out of range

Can you please help it out