v0.4.5

What's Changed

Updates:

• Modularisation of SwyftModule
• Restructuring of configure_optimizers and configure_callbacks
• New default is AdamW with early stopping
• Alternatives with AdamW and OneCycleLR and ReduceLROnPlateau exist as well
• Automatic reloading of best model at the end of training is default
• Extended plot options
• Add smooth_prior flag for interpolating likelihoods
• Renaming of plotting routines (plot_posterior, plot_corner, plot_pair)
• Update example notebook
• Bug fix for labels in corner plot by @anchal-009 in #136

New Contributors

• @anchal-009 made their first contribution in #136
• Co-authored-by: NoemiAM [email protected]

Full Changelog: v0.4.4...v0.4.5

v0.4.4

Overview

• Included updated author list (AUTHOR file) and update LICENSE information
• Add torch conjugate gradient to dependencies (for forthcoming applications)
• Add list of Swyft publications and events to docs
• pytorch-lighthing 2.0 compatibility not fully established (pending until 2.0 has feature parity with 1.9)
• Add swyft.get_class_probs for extracting classification results from ratio estimators
• Add LogRatioEstimator_Gaussian, a Gaussian ratio estimator for arbitrary dimensions
• Deprecate LogRatioEstimator_1dim_Gaussian (worked only with 1dim features and 1dim parameters)
• Rewritten logratios aggregator in swyft.Module to be more resilient
• Add AuxLoss, which just enables to add arbitrary extra losses (e.g. related to regularisation) to the training object
• Enable Spectral embedding for parameters in LogRatioEstimator_Ndim for high fidelity posteriors
• Add LogRatioEstimator_Autoregressive, an autoregressive neural ratio estimator module

What's Changed

• Documentation updates (citations, changelog) by @cweniger in #125
• Update setup.cfg - pytorch-lightning <=1.9.5 by @cweniger in #126
• Merge dev branch by @cweniger in #128
• Updates AUTHORS and LICENSE files by @cweniger in #129
• remove ns by @NoemiAM in #131
• Release/v0.4.4 by @cweniger in #134

Full Changelog: v0.4.3...v0.4.4

v0.4.3

What's Changed

• Added and updated tutorials @cweniger in #122
• Improving documentation @cweniger in #123
• Solving problems with pytorch lightning 2.0.x @cweniger in #124

Full Changelog: v0.4.2...v0.4.3

Release v0.4.2

What's Changed

• Fix plot_zz and plot_pp to deal with N-dim posteriors by @cweniger in #109
• Update README.rst by @bkmi in #114
• Merge current dev branch by @cweniger in #115
• Update setup.cfg by @cweniger in #118
• Bug fix for ZarrStore dataloaders split by @NoemiAM in #120
• Update README.rst by @cweniger in #121

New Contributors

• @NoemiAM made their first contribution in #120

Full Changelog: v0.4.1...v0.4.2

Pre-release v0.4.2

Overview

Fixes:

• Math in docstrings
• Functionality of LogRatioEstimator_1dim_Gaussian, added minstd argument for stability
• Fix tqdm progress bars in notebooks

Enhancements:

• Automatic plot axes assignment
• LogRatioEstimators can take different summaries for different ratios
• Coverage estimation uses noisy logratios in order to make coverage plots useful for classification problems

What's Changed

• Fix plot_zz and plot_pp to deal with N-dim posteriors by @cweniger in #109
• Update README.rst by @bkmi in #114
• Merge current dev branch by @cweniger in #115

Full Changelog: v0.4.1.post1...v0.4.2-pre1

Post-release v0.4.1

Fix image link in readme.rst.

Bugfix release v0.4.1

Fixing installation and documentation bugs

Full Changelog: v0.4.0...v0.4.1

Release v0.4.0 (with new pytorch-lightning based API)

Release of Swyft with pytorch-lightning based API.

What's changed

• Training is handled by pytorch-lightning
• More flexible inference network architectures (enabling training of conditional ratios, ratios of different variables, etc)
• New simulation model based on definition of the full graphical model
• New ZarrStore for parallel simulations
• More efficient handling of coverage tests

See tutorial notebooks for example use-cases.

What's next

• Incorporation of all Swyft v0.3.x functionality (violin plots, etc)

Full Changelog: v0.3.2...v0.4.0

Pre-release v0.4.0 (aka swift-lightning)

First prerelease version of swyft v0.4.0 (aka swyft-lightning) with complete interface overhaul.

swyft: Truncated Marginal Neural Ratio Estimation in Python

• Introduced a version using import swyft; swyft.__version__
• Added labels for violin marginals.
• Followed the requested changes from the JOSS review thread.

Full Changelog: v0.3.1...v0.3.2