Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Save config to hparams.yaml for Tensorboard tracker

See original GitHub issue

Right now, it seems config is ignored when you accelerator.init_trackers(run, config) for Tensorboard tracker. Would be nice if the config is saved as hparams.yaml so it shows up in Tensorboard.

Issue Analytics

State:
Created a year ago
Comments:7 (4 by maintainers)

Top GitHub Comments

2reactions

Shreyz-maxcommented, Aug 18, 2022

@muellerzr cool, thanks for the clarification. I’ll complete it then.

0reactions

muellerzrcommented, Aug 18, 2022

@Shreyz-max that does seem like that’d be the way to do it then, so long as it can be shown in tensorboard correctly

Top Results From Across the Web

Hyperparameter Tuning with the HParams Dashboard

Optimizer. List the values to try, and log an experiment configuration to TensorBoard. This step is optional: you can provide domain information ...

How to save all args to tensorboard and hparams.yaml file?

Hi I would like to save not only the model hparams (the self.save_hypterparameters() method in LightningModule class only saves the model ...

User Guides — Ray 2.2.0 - the Ray documentation

In this section, you can find material on how to use Tune and its various features. You can follow our Tune Feature Guides,...

Use HParams and YAML to Better Manage Hyperparameters ...

To use YAML configs in your python code, we need the class HParams defined in Tensorflow 1.4 API. A HParams object holds hyperparameters ......

Changelog — PyTorch Lightning 1.8.5.post0 documentation

SaveConfigCallback instances should only save the config once to allow ... type hyperparameters to be saved in the haprams.yaml file by TensorBoard and...