Bot releases are hidden (Show)
Change log
_chem_comp.type
logic case-insensitive for mmCIF parsing.Published by joshabramson almost 2 years ago
Version v2.3.1 combines a few small updates.
Change log
Published by Augustin-Zidek almost 2 years ago
Version v2.3.0 updates the AlphaFold-Multimer model parameters. These new models are expected to be more accurate on large protein complexes but use the same model architecture and training methodology as our previously released AlphaFold-Multimer paper. See the v2.3.0 release note for more details.
Thanks to various memory optimisations, AlphaFold-Multimer now uses less GPU memory and it can therefore handle longer proteins.
A number of other bug fixes and small improvements have been made.
Change log
bf16
in multimer inference – reduces GPU memory usage.fp32
when using bf16
in LayerNorm
and replace hk.LayerNorm
with common_modules.LayerNorm
.TriangleMultiplication
to use fused projections and various other memory optimisations.relax_metrics.json
.Published by Htomlinson14 about 2 years ago
Version v2.2.4 is a bug fix release
Change log
jnp.take
to account for behaviour with the new jax version, see https://github.com/deepmind/alphafold/issues/513 (thanks @sokrypton).backbone_loss
, see https://github.com/deepmind/alphafold/issues/570 (thanks @sokrypton).Published by Augustin-Zidek about 2 years ago
Version v2.2.3 is a bug fix release.
Change log
Published by Augustin-Zidek over 2 years ago
A small bug fix release that fixes a bug introduced in v2.2.1 (thanks @lucajovine).
Published by Augustin-Zidek over 2 years ago
Version v2.2.1 is a bug fix release.
Change log
jax.tree_multimap
deprecation warning.run_alphafold_test
(thanks @branfosj).setup.py
(thanks @cmeesters).Published by Augustin-Zidek over 2 years ago
Version v2.2.0 updates the AlphaFold-Multimer model parameters. These new models have greatly reduced numbers of clashes on average and are slightly more accurate. Read the updated AlphaFold-Multimer paper for more details.
A number of other bug fixes and small improvements have been made.
Change log
runtime=nvidia
to expose GPUs to the container (thanks @aburger).num_recycle=0
with models trained with recycling due to incorrect skipping of layers (thanks @sokrypton).split_rng=False
(current default) to sharded_map to support new Haiku release.Published by Augustin-Zidek over 2 years ago
Version v2.1.2 is a bug fix release that also includes the earlier license change.
Change log
enable_gpu_relax
flag (thanks @ojcharles).run_relax
flag (thanks @bkpoon).docker_user
flag (thanks @akors).Published by Augustin-Zidek almost 3 years ago
Version v2.1.1 is a bug fix release for the AlphaFold-Multimer release (v2.1.0).
Change log:
Published by Augustin-Zidek almost 3 years ago
Version 2.1.0 adds the AlphaFold-Multimer model and fixes a number of issues reported in the last few months.
Change log:
model_preset=multimer
to run with AlphaFold-Multimer./tmp
, instead use $TMPDIR
(thanks @meson800, @EricDeveaud).data_dir
, docker_image_name
, output_dir
(thanks @akors, @chrisroat).Published by Augustin-Zidek about 3 years ago
Version 2.0.1 is mainly a bug fix release. We thank everyone who reported issues and proposed solutions.
Change log:
reduced_dbs
preset was added together with small BFD.data/__init__.py
and model/tf/__init__.py
files.tensorflow-cpu
since we don't need a GPU when running the data pipeline.Published by saran-t over 3 years ago