v1.0.11

What's Changed

• Compare motifs by @avantikalal in #162
• added attributions to compare_motifs by @avantikalal in #168
• Transform ensembles by @avantikalal in #111
• Fix wandb login, set anonymous to must by @jkanche in #169

Full Changelog: v1.0.10...v1.0.11

v1.0.10

What's Changed

• updated modiscolite version by @avantikalal in #161
• fixed filter_overlapping with anndata input by @avantikalal in #160
• fixed issue with gc matching by @avantikalal in #159
• Update to the JASPAR 2024 consensus motifs by @oriolfornes in #163

New Contributors

• @oriolfornes made their first contribution in #163

Full Changelog: 1.0.9...v1.0.10

1.0.9

What's Changed

• better-defaults-borzoi-activation-norm by @MuhammedHasan in #153

Full Changelog: v1.0.8...1.0.9

v1.0.8

What's Changed

• Update urls to public gReLU by @jkanche in #146
• Limit ledidi version by @avantikalal in #148
• fix error when return_seqs="best" by @cdedonno in #131
• added hocomoco v13 by @avantikalal in #151
• Tangermeme update by @avantikalal in #152
• fixed ledidi code to work with latest version and reran design tutorial by @avantikalal in #150
• custom fasta by @MuhammedHasan in #143

New Contributors

• @cdedonno made their first contribution in #131

Full Changelog: v1.0.7...v1.0.8

v1.0.7

What's Changed

• allow poisson multinomial loss by @avantikalal in #142
• load catlas dataset from zoo by @avantikalal in #141
• Fixed small Borzoi discrepancies, fixes #144 by @johahi in #147

New Contributors

• @johahi made their first contribution in #147

Full Changelog: v1.0.6...v1.0.7

v1.0.6

What's Changed

• Install tox in pypi-publish.yml by @avantikalal in #120
• set abs=False in saliency by @avantikalal in #124
• Typo in correct_grad causing it to not be applied by @suragnair in #126
• Zoo tutorial by @avantikalal in #122
• changed augmentation and VEP logic by @avantikalal in #121
• split modisco functions by @avantikalal in #128
• created simulation module and tutorial by @avantikalal in #118
• Simulations by @avantikalal in #129
• 🐞 Replace positional arguments with keyword arguments to avoid misassignment of blacklist parameter by @fahimfarhan in #134
• Run fimo on multiple sequences by @jkanche in #136
• fix bug in scan_sequences in motifs.py by @ketringjoni in #137
• updated authors list by @avantikalal in #139
• updated test scan sequences by @avantikalal in #138

New Contributors

• @fahimfarhan made their first contribution in #134
• @ketringjoni made their first contribution in #137

Full Changelog: v1.0.5...v1.0.6

v1.0.5

What's Changed

• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #97
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #100
• print a warning when random mutation augmentation is used by @avantikalal in #101
• add gradient accumulation and checkpoint versioning by @avantikalal in #96
• Prevent padded intervals from exceeding chromosome ends by @avantikalal in #95
• module docstrings by @avantikalal in #104
• corrected use of attr if no motif matches are found by @dagarfield in #99
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #108
• made model checkpoints more informative by @avantikalal in #107
• Revert "made model checkpoints more informative" by @avantikalal in #109
• Make model checkpoints more informative (this time without logging the extra info) by @avantikalal in #114
• added gradient correction by @avantikalal in #106
• Updated contribution guide by @avantikalal in #103
• added a tutorial on custom models by @avantikalal in #102

New Contributors

• @dagarfield made their first contribution in #99

Full Changelog: v1.0.4...v1.0.5

v1.0.4

What's Changed

• Use tangermeme backend for tomtom by @avantikalal in #75
• Bugfix by @avantikalal in #81
• updated available motif sets by @avantikalal in #83
• Score FIMO hits with attributions by @avantikalal in #73
• Fix CI failure by @avantikalal in #90
• fix bug in ism_predict on multiple sequences by @avantikalal in #89

Full Changelog: v1.0.3...v1.0.4

v1.0.3

What's Changed

• fixed bug in pypi-publish workflow by @avantikalal in #35
• Poisson losses bug fixes #31 by @MuhammedHasan in #32
• add python gitignore by @MuhammedHasan in #25
• Add support for half precision (bf16 and fp16) and on device model creation by @MuhammedHasan in #26
• replaced .A with .toarray() by @avantikalal in #37
• allow custom blacklist file by @avantikalal in #36
• replace bpnetlite with tangermeme by @avantikalal in #15
• updated pre-commit hooks by @ekageyama in #44
• Support alternative normalization methods by @MuhammedHasan in #42
• added preprint link by @avantikalal in #49
• Pin torchmetrics version by @gokceneraslan in #51
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #53
• Add precision to the pl.Trainer constructors by @gokceneraslan in #50
• fixed and updated plot_tracks by @avantikalal in #67
• Saliency by @avantikalal in #71
• Flash Attention support by @suragnair in #78
• updated merge_intervals function to account for strand by @avantikalal in #66
• Updated motif processing to Tangermeme 0.4 by @avantikalal in #69
• add multiomial axis option to poisson multinomial loss by @avantikalal in #79

New Contributors

• @MuhammedHasan made their first contribution in #32
• @ekageyama made their first contribution in #44
• @pre-commit-ci made their first contribution in #53

Full Changelog: v1.0.2...v1.0.3

v1.0.2

What's Changed

• Reset val/test metrics and losses after every epoch by @suragnair in #8
• updated readme with pip command by @avantikalal in #14
• convert indices to int8 before stacking by @avantikalal in #21
• fixed device parsing in embed_on_dataset by @avantikalal in #20
• added host and docstrings to resources module by @avantikalal in #13
• added map_location arg in load_model to load model onto any device by @avantikalal in #18
• added reset method to PearsonCorrCoef by @avantikalal in #30

New Contributors

• @suragnair made their first contribution in #8

Full Changelog: v1.0.1...v1.0.2