Skip to content
This repository has been archived by the owner on Jun 26, 2021. It is now read-only.

[FeatureRequest] Add logging Frequency during training #48

Closed
ORippler opened this issue Jan 29, 2019 · 1 comment
Closed

[FeatureRequest] Add logging Frequency during training #48

ORippler opened this issue Jan 29, 2019 · 1 comment

Comments

@ORippler
Copy link
Collaborator

Description
Provide means to specify logging Frequency during training. For Visdom, all logging results are kept in memory (risking OOM), and with tensorboardlogger, disk I/O may bottleneck training otherwise.

Proposal
There are different ways to tackle this. One could be to utilize CombinedLogger from trixi. However, I'm unsure whether CombinedLogger allows for logging frequencies assigned per logged object (e.g. Scalars every iteration and images every Nth iteration), which is a feature I would like to have.

Additional context
Solve before implement logging frequencies as part of #47

@ORippler ORippler added this to ToDo in PyTorch via automation Jan 29, 2019
@ORippler ORippler added this to ToDo in Tensorflow via automation Jan 29, 2019
@justusschock
Copy link
Member

I'd like to address this, once #139 is done, since we can then introduce a logging frequency on our own (without having to hack trixi anymore)

@justusschock justusschock self-assigned this Aug 1, 2019
PyTorch automation moved this from ToDo to Done Sep 11, 2019
Tensorflow automation moved this from ToDo to Done Sep 11, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
PyTorch
  
Done
Tensorflow
  
Done
Development

Successfully merging a pull request may close this issue.

2 participants