You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Mar 17, 2021. It is now read-only.
It seems that in many window-based segmentation/regression use cases, either settingis_training=True or setting use_local_stats=True in the batch normalisation layer at inference time leads to better performance than those with NiftyNet's default batch norm options.
Therefore this ticket proposes to add a use_local_stats option in the config file, and make use_local_stats=True as the default for segmentation application with batch norm.
Specifically, in the current implementation (v0.4.0) the default values are:
training: is_trainign=Trueuse_local_stats=True
inference: is_training=Falseuse_local_stats=True
which uses running averages from training at inference time, following Ioffe and Szegedy (2015)https://arxiv.org/abs/1502.03167.
The proposed change will use these values by default:
training: is_trainign=Trueuse_local_stats=True
inference: is_training=Falseuse_local_stats=True
which will effectively follow Ulyanov et al. (2017)https://arxiv.org/abs/1607.08022.
It seems that in many window-based segmentation/regression use cases, either setting
is_training=True
or settinguse_local_stats=True
in the batch normalisation layer at inference time leads to better performance than those with NiftyNet's default batch norm options.Therefore this ticket proposes to add a
use_local_stats
option in the config file, and makeuse_local_stats=True
as the default for segmentation application with batch norm.Specifically, in the current implementation (v0.4.0) the default values are:
is_trainign=True
use_local_stats=True
is_training=False
use_local_stats=True
which uses running averages from training at inference time, following Ioffe and Szegedy (2015) https://arxiv.org/abs/1502.03167.
The proposed change will use these values by default:
is_trainign=True
use_local_stats=True
is_training=False
use_local_stats=True
which will effectively follow Ulyanov et al. (2017) https://arxiv.org/abs/1607.08022.
This ticket is related to #280, #285, #282.
thoughts?
Probably to be addressed with #282
The text was updated successfully, but these errors were encountered: