solo-learn

solo-learn copied to clipboard

Question about --scale_loss

3

Hi Could you teach me the effect of `--scale_loss`? Does the option normalize the gradient of batch normalization and bias parameters if I use it with `--exclude_bias_n_norm`, as explained in...

yutaro-s

Implement multi-crop (SwAV, DCv2, DINO, MoCo)

1

and improve SimCLR's implementation

DonkeyShot21

DonkeyShot21

DonkeyShot21

Separate train and validation datasets without and with labels

2

Hi there! First of all nice work on the library, looks great. I am in the special situation where I have a large dataset for pretraining but without any labels....

nsmdgr

Move from Argparse to Omegaconf

11

Opening the PR now so that we can properly visualize the changes and eventually get the tests rolling.

vturrisi

Is there an e2e integration test on toy data?

6

**Describe the bug** In doing a major refactor (e.g. switching to OmegaConf or hydra), it's not clear to me there is a full e2e integration test. What main(s) would be...

turian

Remove dim=1 parameters from LARS updates and weight decay

8

**Describe the bug** In addition to `solo.utils.misc.remove_bias_and_norm_from_weight_decay`, we might consider also `solo.utils.misc.ignore_dim1_parameters` to ignore weight decay *and `lars_adaptation_filter`* them form. **Versions** solo-learn main **Additional comments** I see that `solo.utils.misc.remove_bias_and_norm_from_weight_decay` was...

turian

vturrisi

[pre-commit.ci] pre-commit autoupdate

1

updates: - [github.com/pre-commit/pre-commit-hooks: v4.5.0 → v4.6.0](https://github.com/pre-commit/pre-commit-hooks/compare/v4.5.0...v4.6.0) - [github.com/kynan/nbstripout: 0.6.1 → 0.7.1](https://github.com/kynan/nbstripout/compare/0.6.1...0.7.1) - [github.com/mwouts/jupytext: v1.16.0 → v1.16.2](https://github.com/mwouts/jupytext/compare/v1.16.0...v1.16.2) - [github.com/psf/black: 23.12.0 → 24.4.2](https://github.com/psf/black/compare/23.12.0...24.4.2)

pre-commit-ci[bot]