search

netw0rkf10w / CRF

Conditional Random Fields
Apache License 2.0
23 stars 2 forks source link

Support for distributed training #7

Closed Ending2015a closed 2 years ago

Ending2015a commented 2 years ago

Hello, thank you for this amazing work!

I want to use this CRF package for distributed training. I wonder if it is OK to use it to train a segmentation model across many GPUs or processes. For example, does it contain any operations like batch normalization that needs to perform computations across batches?

netw0rkf10w commented 2 years ago

Hi @Ending2015a. Thanks for your interest! You can safely use CRF for any kind of distributed training (e.g., Pytorch DDP or DataParallel). I will try to release my semantic segmentation code soon. The code is actually ready but it takes a lot of time to write a good README, which I started but had to stop to work on some deadlines. Please feel free to ask if you have any questions.

Ending2015a commented 2 years ago

Thank you!