graphs: v0.6.3

0.6.3 (2025-08-01)

Features

• graphs: Add lat weighted attribute (#223) (5dd32ca)
• graphs: Support to export edges to npz (#395) (e21738f)

Bug Fixes

• Dropping 3.9 (#436) (f6c0214)
• graphs: Revert PR #379 (#409) (d51219f)
• graphs: Throw error instead of raising warning when graph exists. (#379) (6ec6c18)
• graphs: Undo masking when torch-cluster is installed (#375) (9f75c06)

Documentation

• graphs: Documenting some missing features (#423) (8addbd8)

training: v0.5.1

0.5.1 (2025-06-17)

Features

• models,training: Shard everything (#121) (06dde94)

Bug Fixes

• Config validator checks for conflict with bounding and graphtran… (#352) (f97313e)
• training, losses: NaN handling for crps (#358) (daf9ddd)

models: v0.8.1

0.8.1 (2025-06-17)

Features

• models,training: Shard everything (#121) (06dde94)

Bug Fixes

• Do not inherit from a concrete class (#359) (ca79375)
• Revert PR 359 (#365) (fcebea4)

graphs: v0.6.2

0.6.2 (2025-06-17)

Features

• graphs: Add GridsMask node attribute builder (#355) (c4db823)
• graphs: Build nodes from xarray compatible data (#330) (3edaabb)

Bug Fixes

• graphs: Move libraries to optional dependencies (#338) (db215ce)
• SphericalAreaWeights (#363) (acca570)

training: v0.5.0

0.5.0 (2025-06-05)

⚠ BREAKING CHANGES

• models,training: Remove multimapper (#268)
• generalise activation function (#163)
• Rework Loss Scalings to provide better modularity (#52)

Features

• Generalise activation function (#163) (98c4d06)
• graphs: Add scale_resolutions arg (#188) (0ea0642)
• models,training: Remove multimapper (#268) (0e8bb99)
• Transformer mapper (#179) (2cea7db)

Bug Fixes

• Adapt ensemble configs to work with loss refactor. (#300) (c210478)
• Adapt predict_step in model interface to pass on arguments for model classes (#281) (a5b2643)
• Change output format in validation step to fix issue in plotting (#305) (cc941f3)
• Clean params complement (#291) (4e9ca31)
• Dataset_order (#334) (762227a)
• deps: Bump anemoi-models >= 0.8 (#351) (1fbd525)
• models,training: Remove unnecessary torch-geometric maximum version (#326) (fe93ea8)
• Move scaler to device in scale (#317) (1592d09)
• Remove activation entry from MLP noise block (#340) (2d060f5)
• Rework Loss Scalings to provide better modularity (#52) (162b906)
• training, models: Update interpolator to work with new features (#322) (cfdc99f)
• training: Bump anemoi-graphs version to 0.5.2 (#276) (9b8ec13)
• training: Explicit Batch invariance (#318) (45f6e15)
• training: No logging variable metadata to mlflow (#304) (4f7c2c9)
• Update defaults for mlflow logging (#333) (5560581)

Documentation

• Downloading era5 o96 dataset (#307) (b9e2d24)
• Fix minor mistakes in CRPS user guide. (#264) (52d9a2e)

models: v0.8.0

0.8.0 (2025-06-05)

⚠ BREAKING CHANGES

• models,training: Remove multimapper (#268)

Features

• models,training: Remove multimapper (#268) (0e8bb99)

Bug Fixes

• Dataset_order (#334) (762227a)
• training, models: Update interpolator to work with new features (#322) (cfdc99f)

graphs: v0.6.1

0.6.1 (2025-06-05)

Bug Fixes

• graphs: Fix device mismatch error (#339) (8712c2a)

models: v0.7.0

0.7.0 (2025-05-30)

⚠ BREAKING CHANGES

• generalise activation function (#163)

Features

• generalise activation function (#163) (98c4d06)
• transformer mapper (#179) (2cea7db)

Bug Fixes

• models,training: Remove unnecessary torch-geometric maximum version (#326) (fe93ea8)
• remove activation entry from MLP noise block (#340) (2d060f5)

models: v0.6.0

0.6.0 (2025-05-15)

⚠ BREAKING CHANGES

• Rework Loss Scalings to provide better modularity (#52)

Features

• GraphtransformerProcessor chunking (#66) (1daa9f2)

Bug Fixes

• Rework Loss Scalings to provide better modularity (#52) (162b906)

graphs: v0.6.0

0.6.0 (2025-05-16)

⚠ BREAKING CHANGES

• Rework Loss Scalings to provide better modularity (#52)

Bug Fixes

• graphs,directions: fix error with direction computation (#287) (55feded)
• Rework Loss Scalings to provide better modularity (#52) (162b906)