Update flax requirement from <0.4,>=0.3.5 to >=0.3.5,<0.5
Created by: dependabot[bot]
Updates the requirements on flax to permit the latest version.
Release notes
Sourced from flax's releases.
Version 0.4.0
What's Changed
- Add PReLU Activation by
@isaaccorley
in google/flax#1570- Fix GroupNorm type hint for param num_groups. by
@lkhphuc
in google/flax#1657- Add named_call overrides to docs by
@jheek
in google/flax#1649- mission statement by
@jheek
in google/flax#1668- Improves Flax Modules for RTD by
@marcvanzee
in google/flax#1416- Add clarifying docstring for 'size' argument to prefetch_to_device's by
@avital
in google/flax#1574- Add circular padding to flax.linen.Conv and flax.linen.ConvTranspose by
@sgrigory
in google/flax#1661- Fix child scope rng reuse. by
@jheek
in google/flax#1692- Numerically stable weight norm by
@jheek
in google/flax#1693- Remove cyclic refs from scope by
@jheek
in google/flax#1696- Add
unroll
tojax_utils.scan_in_dim
by@ptigwe
in google/flax#1691- Removes
rng
arguments from Dropout's__call__
. by@copybara-service
in google/flax#1689- Add error for empty scopes. by
@jheek
in google/flax#1698- correct axis resolution in case of repeated axis in the logica axis r… by
@ultrons
in google/flax#1703- Fix lost mutation bug in transforms on nested scopes. by
@levskaya
in google/flax#1716- Expose put_variable function to Module. by
@levskaya
in google/flax#1710- add eq and hash for scopes by
@jheek
in google/flax#1720- Fixes a bug in DenseGeneral. by
@copybara-service
in google/flax#1722- Add param_dtype argument to linen Modules by
@jheek
in google/flax#1739- Implement custom vjp by
@jheek
in google/flax#1738- Handle setup with transformed methods taking submodules of self. by
@levskaya
in google/flax#1745- validate RNG key shape against jax's default by
@copybara-service
in google/flax#1780- Adds optax update guide. by
@andsteing
in google/flax#1774- Implement LazyRNG by
@jheek
in google/flax#1723- make params_with_axes() work when params_axes is not mutable by
@copybara-service
in google/flax#1811- Updates the ensembling HOWTO to Optax. by
@andsteing
in google/flax#1806- Adds prominent
scenic
link toexamples/README.md
by@copybara-service
in google/flax#1809- Removes PixelCNN++ example.
@copybara-service
in google/flax#1819- Add support for non-float32 normalization for linen normalization layers by
@jheek
in google/flax#1804- Make Filter a Collection instead of a Container by
@NeilGirdhar
in google/flax#1815- Removes deprecated API from RTD by
@marcvanzee
in google/flax#1824New Contributors
@isaaccorley
made their first contribution in google/flax#1570@lkhphuc
made their first contribution in google/flax#1657@sgrigory
made their first contribution in google/flax#1661@ptigwe
made their first contribution in google/flax#1691@ultrons
made their first contribution in google/flax#1703@dependabot
made their first contribution in google/flax#1749@NeilGirdhar
made their first contribution in google/flax#1699@saeta
made their first contribution in google/flax#1784@melissatan
made their first contribution in google/flax#1793Full Changelog: https://github.com/google/flax/compare/v0.3.6...v0.4.0
Changelog
Sourced from flax's changelog.
0.4.0
Breaking changes:
- flax.deprecated.nn is removed. Please pin to flax==0.3.6 if you are still using it.
- PixelCNN++ example is removed. It was not working well on TPU.
- linen Normalization layers no longer downcast double and complex floats tofloat32 when computing the mean and variance.
New features:
- Added
flax.linen.custom_vjp
for custom derivatives inside aModule
.- Add
param_dtype
attribute to standard Linen Modules for specifying parameter dtypes.0.3.6
Breaking changes:
- Move
flax.nn
toflax.deprecated.nn
.New features:
- Add experimental checkpoint policy argument. See
flax.linen.checkpoint
- Add lifted versions of jvp and vjp.
- Add lifted transformation for mapping variables. See
flax.linen.map_variables
.0.3.5
Breaking changes:
- You can no longer pass an int as the
kernel_size
for a `flax.linen.Conv. Instead a type error is raised stating that a tuple/list should be provided. Stride and dilation arguments do support broadcasting a single int value now because this is not ambigious when the kernel rank is known.flax.linen.enable_named_call
andflax.linen.disable_named_call
now work anywhere instead of only affecting Modules constructed after the enable/disable call. Additionally, there is nowflax.linen.override_named_call
that provided a context manager to locally disable/enable named_call.- NamedTuples are no longer converted to tuples on assignment to a
linen.Module
.New features:
- Flax internal stack frames are now removed from exception state traces.
- Added
flax.linen.nowrap
to decorate method that should not be transformed because they are stateful.- Flax no longer uses implicit rank broadcasting. Thus, you can now use Flax with
--jax_numpy_rank_promotion=raise
.Bugfixes:
- linen Modules and dataclasses made with
flax.struct.dataclass
orflax.struct.PyTreeNode
are now correctly recognized as dataclasses by static analysis tools like PyLance. Autocomplete of constructors has been verified to work with VSCode.- Fixed a bug in FrozenDict which didn't allow copying dicts with reserved names.
- Fix the serialization of named tuples. Tuple fields are no longer stored in the state dict and the named tuple class is no longer recreated (bug).
- Mixed precision training with float16 now works correctly with the attention layers.
- auto-generated linen Module
__hash__
,__eq__
,__repr__
no longer fail by default on non-init attributes.
... (truncated)
Commits
-
78d0566
Merge pull request #1825 from marcvanzee:pipy -
695a880
Merge branch 'main' of https://www.github.com/google/flax into pipy -
5de87fa
Bump Flax to v0.4.0 -
a6e7744
Merge pull request #1824 from marcvanzee:rtd-deprecated -
a7cf071
Updates Linen README -
1c4fe91
Updates core notebook to Linen -
12a1159
Removes deprecated API from docs -
aca40ad
Actually removes the deprecated folder. -
948ee26
Removes flax.deprecated.nn. -
7288ada
Merge pull request #1815 from NeilGirdhar:container - Additional commits viewable in compare view
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase
.
Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
-
@dependabot rebase
will rebase this PR -
@dependabot recreate
will recreate this PR, overwriting any edits that have been made to it -
@dependabot merge
will merge this PR after your CI passes on it -
@dependabot squash and merge
will squash and merge this PR after your CI passes on it -
@dependabot cancel merge
will cancel a previously requested merge and block automerging -
@dependabot reopen
will reopen this PR if it is closed -
@dependabot close
will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually -
@dependabot ignore this major version
will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) -
@dependabot ignore this minor version
will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) -
@dependabot ignore this dependency
will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)