chore(deps): update dependency lightly to v1.5.26

[renovate]

This PR contains the following updates:

Package	Change	Age	Confidence
lightly	==1.5.22 → ==1.5.26

---

Release Notes

lightly-ai/lightly (lightly)

v1.5.26

Compare Source

What's Changed

• docs: align SimCLR transform docstrings with Google style by @​saud5150 in #​1948
• docs: add docstrings to BYOLView1Transform and BYOLView2Transform by @​saud5150 in #​1946
• docs: add docstrings to DINOViewTransform and IBOTViewTransform by @​saud5150 in #​1950
• docs: add docstrings to MSN, SimSiam, VICReg, and VICRegL view transforms by @​saud5150 in #​1951
• chore: add codespell pre-commit hook and fix typos by @​Borda in #​1952
• docs: add docstrings to SwaVViewTransform and SmoGViewTransform by @​saud5150 in #​1954
• chore: add mypy to pre-commit by @​saud5150 in #​1947
• docs: add docstrings to FDAView1Transform, FDAView2Transform, and DetConSViewTransform by @​saud5150 in #​1955
• Add PatchKernelAlignmentLoss (PaKa paper) by @​IgorSusmelj in #​1953
• examples: add FDA training example (pytorch) by @​saud5150 in #​1959
• examples: add FDA training examples (pytorch_lightning, pytorch_lightning_distributed) by @​saud5150 in #​1964
• Add DetConS pytorch training example by @​saud5150 in #​1961
• Add FroSSL by @​jacobsn in #​1962
• examples: add scikit-image segmentation path to DetConS pytorch example by @​saud5150 in #​1969
• examples: add DetConS training examples (pytorch_lightning, pytorch_lightning_distributed) by @​saud5150 in #​1971
• test(models): loosen tolerance in test_singular_mask for FP summation noise by @​gabrielfruet in #​1970
• Add VISRegLoss (VISReg paper) by @​saud5150 in #​1968
• docs: add DetConS documentation page and README entry by @​saud5150 in #​1973
• Remove Codecov integration by @​yashkuceriya in #​1967
• Add Pixio by @​lorinczszabolcs in #​1965
• docs: sort model table by release date instead of alphabetically by @​minagayid in #​1975
• fix(loss): use autograd-aware all_reduce in BarlowTwinsLoss by @​saud5150 in #​1980
• Add CLAUDE.md for AI assistant guidance by @​gabrielfruet in #​1984
• feat(transforms): add mask_fn to DetConSTransform for callable segmentation by @​saud5150 in #​1981
• docs: add Pixio to README methods table by @​lorinczszabolcs in #​1986
• Pre-release v1.5.26 by @​gabrielfruet in #​2003

New Contributors

• @​Borda made their first contribution in #​1952
• @​jacobsn made their first contribution in #​1962
• @​yashkuceriya made their first contribution in #​1967
• @​lorinczszabolcs made their first contribution in #​1965
• @​minagayid made their first contribution in #​1975

Full Changelog: lightly-ai/lightly@v1.5.25...v1.5.26

v1.5.25

Compare Source

What's Changed

• Add LeJEPA ViT-S/16 benchmark results by @​gabrielfruet in #​1935
• Fix: ibotpluplus loss normalisation by @​IgorSusmelj in #​1937
• fix(loss): raise clear error for single-view DINOLoss instead of silent NaN by @​devteamaegis in #​1938
• docs: make LeJEPA public-facing (TRN-2158) by @​gabrielfruet in #​1940
• docs: add docstring to SimCLRViewTransform by @​saud5150 in #​1943
• fix: return plain Python float from cosine_schedule by @​saud5150 in #​1941
• Release: bump version to 1.5.25 by @​gabrielfruet in #​1944

New Contributors

• @​devteamaegis made their first contribution in #​1938
• @​saud5150 made their first contribution in #​1943

Full Changelog: lightly-ai/lightly@v1.5.24...v1.5.25

v1.5.24

Compare Source

Implement LeJEPA!

What's Changed

• isort and black for python pre-commit hook by @​gabrielfruet in #​1909
• [DOCS] Fix errors in CONTRIBUTING.md by @​RecreationalMath in #​1911
• [MNT-style]: Migrate from black, isort, and pylint to ruff by @​RecreationalMath in #​1914
• Main, knn_classifier and simclr modifications by @​Quentincami in #​1901
• fix override errors in masked causal vision transformer by @​gabrielfruet in #​1913
• Add more Lightly in Research papers by @​gabrielfruet in #​1921
• [MNT-style]: Add ruff pydocstyle for Google-style docstrings by @​RecreationalMath in #​1918
• Fix SIGReg distributed gradient under DDP by @​RecreationalMath in #​1923
• [ENH] Add LeJEPA core building blocks by @​RecreationalMath in #​1916
• refactor: simplify LeJEPA loss and module surface by @​gabrielfruet in #​1926
• [MNT] SIGReg cleanup: dedup tests, backfill docstring, document weights by @​RecreationalMath in #​1925
• Add LEJEPA examples by @​gabrielfruet in #​1927
• feat: Implement DSE (Dense representation Structure Estimator) by @​IgorSusmelj in #​1929
• docs: consolidate imagenet benchmark readme by @​gabrielfruet in #​1931
• feat: implement iBOT++ loss by @​IgorSusmelj in #​1930
• Release LightlySSL v1.5.24 by @​gabrielfruet in #​1933
• feat: add LeJEPA benchmark and metrics by @​gabrielfruet in #​1932

New Contributors

• @​RecreationalMath made their first contribution in #​1911
• @​Quentincami made their first contribution in #​1901

Full Changelog: lightly-ai/lightly@v1.5.23...v1.5.24

v1.5.23

Compare Source

What's Changed

• Fix deprecation warning totensor (#​1864) by @​KylevdLangemheen in #​1866
• Move functional transforms API to compatibility layer (#​1867) by @​KylevdLangemheen in #​1870
• Fix the banner and tab name of LightlySSL Docs by @​yutong-xiang-97 in #​1872
• Add DINOv2 to Readme (#​1873) by @​KylevdLangemheen in #​1875
• Fixed SMoG examples by @​thomasmarchioro3 in #​1876
• Add DirectCLR (#​781) by @​KylevdLangemheen in #​1874
• Update README badge links by @​Olexandr88 in #​1880
• Add link to social network twitter by @​Olexandr88 in #​1883
• Update packages for CI by @​IgorSusmelj in #​1896
• Contributing Guide: Add instructions on Imports by @​liopeer in #​1895
• Add LeJEPA SIGReg Loss implementation by @​gabrielfruet in #​1893
• Speed up momentum update by @​vsey in #​1899
• Add LightlyStudio reference to README by @​michal-lightly in #​1903
• Add DINOv2-3D: Self-Supervised 3D Vision Transformer Pretraining to README by @​guarin in #​1890
• Add torchvision 0.26 support by @​guarin in #​1904

New Contributors

• @​thomasmarchioro3 made their first contribution in #​1876
• @​Olexandr88 made their first contribution in #​1880
• @​gabrielfruet made their first contribution in #​1893
• @​vsey made their first contribution in #​1899

Full Changelog: lightly-ai/lightly@v1.5.22...v1.5.23

Many thanks to our contributors!

Models

• AIM: Scalable Pre-training of Large Autoregressive Image Models
• Barlow Twins: Self-Supervised Learning via Redundancy Reduction, 2021
• Bootstrap your own latent: A new approach to self-supervised Learning, 2020
• DCL: Decoupled Contrastive Learning, 2021
• DenseCL: Dense Contrastive Learning for Self-Supervised Visual Pre-Training, 2021
• DINO: Emerging Properties in Self-Supervised Vision Transformers, 2021
• DINOv2: Learning Robust Visual Features without Supervision, 2023
• FastSiam: Resource-Efficient Self-supervised Learning on a Single GPU, 2022
• iBOT: Image BERT Pre-Training with Online Tokenizer, 2021
• I-JEPA: Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture, 2023
• MAE: Masked Autoencoders Are Scalable Vision Learners, 2021
• MSN: Masked Siamese Networks for Label-Efficient Learning, 2022
• MoCo: Momentum Contrast for Unsupervised Visual Representation Learning, 2019
• NNCLR: Nearest-Neighbor Contrastive Learning of Visual Representations, 2021
• PMSN: Prior Matching for Siamese Networks, 2022
• SimCLR: A Simple Framework for Contrastive Learning of Visual Representations, 2020
• SimMIM: A Simple Framework for Masked Image Modeling, 2021
• SimSiam: Exploring Simple Siamese Representation Learning, 2020
• SMoG: Unsupervised Visual Representation Learning by Synchronous Momentum Grouping, 2022
• SwAV: Unsupervised Learning of Visual Features by Contrasting Cluster Assignments, M. Caron, 2020
• TiCo: Transformation Invariance and Covariance Contrast for Self-Supervised Visual Representation Learning, 2022
• VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning, Bardes, A. et. al, 2022
• VICRegL: VICRegL: Self-Supervised Learning of Local Visual Features, 2022

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • At any time (no schedule defined)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.

[renovate]

Renovate Ignore Notification

Because you closed this PR without merging, Renovate will ignore this update (==1.5.26). You will get a PR once a newer version is released. To ignore this dependency forever, add it to the ignoreDeps array of your Renovate config.

If you accidentally closed this PR, or if you changed your mind: rename this PR to get a fresh replacement PR.