chore(main): Preparing Next Release for anemoi-models 0.4.2

[DeployDuck]

🤖 Automated Release PR

This PR was created by release-please to prepare the next release. Once merged:

1. A new version tag will be created
2. A GitHub release will be published
3. The changelog will be updated

Changes to be included in the next release:

0.4.2 (2025-02-04)

What's Changed

Training

• fix(training, plots) Exclude nans from error colorbars by @anaprietonem in fix(training, plots) Exclude nans from error colorbars #59
• fix(training): bump anemoi-datasets required version to 0.5.13 by @JPXKQX in fix(training): bump anemoi-datasets required version to 0.5.13 #74
• chore(training): Add default config files for 2 and 3 level hierarchical processors by @JPXKQX in chore(training): Add default config files for 2 and 3 level hierarchical processors #50
• fix: update graph configs to avoid DeprecationWarning for area weights by @JPXKQX in fix: update graph configs to avoid DeprecationWarning for area weights #53
• feat(models): normalization layers by @jakob-schloer in feat(models): normalization layers #47
• Fix crash in diagnostic plots (colorbar limits) by @lzampier in Fix crash in diagnostic plots (colorbar limits) #85
• fix(training): profiler 'Model Summary' works when sharding models over multiple GPUs by @cathalobrien in fix(training): profiler 'Model Summary' works when sharding models over multiple GPUs #90
• docs(graphs): Refactor anemoi-graphs documentation by @JPXKQX in docs(graphs): Refactor anemoi-graphs documentation #49
• fix: pin dask version to 2024.12.1 by @JPXKQX in fix: pin dask version to 2024.12.1  #94
• docs: Improve installation docs by @HCookie in docs: Improve installation docs #91
• docs: cancel RTD builds on PRs without change by @JesperDramsch in docs: cancel RTD builds on PRs without change #97
• feat: Model Freezing ❄️ by @icedoom888 in feat: Model Freezing ❄️  #61
• feat: make flash attention configurable by @theissenhelen in feat: make flash attention configurable #60
• fix: cpu memory savings of sharded dataloader by @japols in fix: cpu memory savings of sharded dataloader #83
• chore: synced file(s) with ecmwf-actions/reusable-workflows by @DeployDuck in chore: synced file(s) with ecmwf-actions/reusable-workflows #84

Models

• feature(models): Add model comm group to predict_step by @cathalobrien in feature(models): Add model comm group to predict_step  #77
• Implementation of NormalizedReluBounding for non-zero thresholds by @lzampier in Implementation of NormalizedReluBounding for non-zero thresholds #64
• fix: normalise in place to reduce memory by @japols in fix: normalise in place to reduce memory #82
• feat(models): use num_layers of the processor in hierarchical graphs by @JPXKQX in feat(models): use num_layers of the processor in hierarchical graphs #78
• fix: default behaviour for kernel_layers when not set in config. by @jakob-schloer in fix: default behaviour for kernel_layers when not set in config. #93
• fix: bug in variables ordering in NormalizedReluBounding by @lzampier in fix: bug in variables ordering in NormalizedReluBounding #98
• feat(models): Copy Imputer by @icedoom888 in feat(models): Copy Imputer #72

Graphs

• feat(graphs,plots): expand support for multi-dimensional node attributes by @JPXKQX in feat(graphs,plots): expand support for multi-dimensional node attributes #48
• feat(graphs): New Edge Attribute: AttributeFromNode by @icedoom888 in feat(graphs): New Edge Attribute: AttributeFromNode #62
• feat: support ReducedGaussianGridNodes by @JPXKQX in feat: support ReducedGaussianGridNodes #54
• chore(main): Preparing Next Release for anemoi-graphs 0.4.3 by @DeployDuck in chore(main): Preparing Next Release for anemoi-graphs 0.4.3 #110
• chore(main): Preparing Next Release for anemoi-graphs 0.4.4 by @DeployDuck in chore(main): Preparing Next Release for anemoi-graphs 0.4.4 #111

Other Changes

• pre-commits-for-models-graphs-dev by @sahahner in pre-commits-for-models-graphs-dev #45
• ci(docs): bring ReadTheDocs CI pipeline by @JPXKQX in ci(docs): bring ReadTheDocs CI pipeline #73
• ci: Reinstantiate CI files by @JesperDramsch in ci: Reinstantiate CI files #75
• ci: Propose release-please implementation by @JesperDramsch in ci: Propose release-please implementation #100

New Contributors

• @anaprietonem made their first contribution in fix(training, plots) Exclude nans from error colorbars #59
• @cathalobrien made their first contribution in feature(models): Add model comm group to predict_step  #77
• @lzampier made their first contribution in Implementation of NormalizedReluBounding for non-zero thresholds #64
• @japols made their first contribution in fix: normalise in place to reduce memory #82
• @JesperDramsch made their first contribution in ci: Reinstantiate CI files #75
• @jakob-schloer made their first contribution in feat(models): normalization layers #47
• @icedoom888 made their first contribution in feat: Model Freezing ❄️  #61
• @theissenhelen made their first contribution in feat: make flash attention configurable #60
• @DeployDuck made their first contribution in chore: synced file(s) with ecmwf-actions/reusable-workflows #84

Full Changelog: anemoi-models-0.4.1...anemoi-models-0.4.2

---

Important

⚠️ Merging this PR will:

• Create a new release
• Trigger deployment pipelines
• Update package versions

Before merging:

• Ensure all tests pass
• Review the changelog carefully
• Get required approvals

Release-please documentation