lhoyer / DAFormer

[CVPR22] Official Implementation of DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation
Other
466 stars 91 forks source link

Save intermediate weight, and resume training #80

Closed ShenZheng2000 closed 1 year ago

ShenZheng2000 commented 1 year ago

The current version of the program only saves the final weight, which can be inconvenient if the training is interrupted in the middle, and there's a need to resume training from a specific point.

Is there a way to save intermediate weights at fixed iterations during training, allowing us to resume training from those checkpoints if needed?