Skip to content

Add support for many additional Keras 3 layers #465

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Dobiasd merged 9 commits into from
May 3, 2026
Merged

Add support for many additional Keras 3 layers #465

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
2309fdd
Add support for many additional Keras 3 layers
Dobiasd May 1, 2026
8631738
Apply auto_format.sh
Dobiasd May 1, 2026
b824708
Revert attention_layer.hpp formatting change
Dobiasd May 1, 2026
ea915b3
Address review findings
Dobiasd May 1, 2026
2ce984e
Split test_model_recurrent into recurrent + extended
Dobiasd May 2, 2026
41afc71
Move newly-added non-recurrent layers into test_model_exhaustive
Dobiasd May 2, 2026
baf492d
Refactor adaptive_pooling, einsum_dense, GQA for readability
Dobiasd May 2, 2026
568321e
Address second-round review findings
Dobiasd May 2, 2026
a9ea31e
Add regression test for EinsumDense bias_axes ordering
Dobiasd May 2, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 21 additions & 15 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,21 +41,31 @@ Would you like to build/train a model using Keras/Python? And would you like to

* `Add`, `Concatenate`, `Subtract`, `Multiply`, `Average`, `Maximum`, `Minimum`, `Dot`
* `AveragePooling1D/2D/3D`, `GlobalAveragePooling1D/2D/3D`
* `AdaptiveAveragePooling1D/2D/3D`, `AdaptiveMaxPooling1D/2D/3D`
* `TimeDistributed`
* `Conv1D/2D/3D`, `SeparableConv2D`, `DepthwiseConv2D`
* `Conv1D/2D/3D`, `SeparableConv2D`, `DepthwiseConv1D`, `DepthwiseConv2D`
* `Conv1DTranspose`, `Conv2DTranspose`, `Conv3DTranspose`
* `Cropping1D/2D/3D`, `ZeroPadding1D/2D/3D`, `CenterCrop`
* `BatchNormalization`, `Dense`, `Flatten`, `Normalization`
* `BatchNormalization`, `Dense`, `EinsumDense`, `Flatten`, `Normalization`
* `Dropout`, `AlphaDropout`, `GaussianDropout`, `GaussianNoise`
* `SpatialDropout1D`, `SpatialDropout2D`, `SpatialDropout3D`
* `ActivityRegularization`, `LayerNormalization`, `UnitNormalization`
* `RandomContrast`, `RandomFlip`, `RandomHeight`
* `RandomRotation`, `RandomTranslation`, `RandomWidth`, `RandomZoom`
* `ActivityRegularization`, `LayerNormalization`, `RMSNormalization`
* `GroupNormalization`, `UnitNormalization`
* Training-only image augmentation layers (passed through at inference):
`RandomBrightness`, `RandomContrast`, `RandomCrop`, `RandomFlip`, `RandomHue`,
`RandomGrayscale`, `RandomRotation`, `RandomTranslation`, `RandomZoom`,
`RandomShear`, `RandomSaturation`, `RandomPerspective`, `AutoContrast`,
`AugMix`, `CutMix`, `MixUp`, `RandAugment`, `Solarization`, `Equalization`
* `MaxPooling1D/2D/3D`, `GlobalMaxPooling1D/2D/3D`
* `UpSampling1D/2D/3D`, `Resizing`, `Rescaling`
* `Reshape`, `Permute`, `RepeatVector`
* `Embedding`, `CategoryEncoding`
* `Attention`, `AdditiveAttention`, `MultiHeadAttention`
* `Discretization`, `IntegerLookup`
* `Masking` (passthrough at inference)
* `Attention`, `AdditiveAttention`, `MultiHeadAttention`, `GroupedQueryAttention`
* `LSTM`, `GRU`, `SimpleRNN`, `Bidirectional`
* `RNN` wrapping `LSTMCell`/`GRUCell`/`SimpleRNNCell`/`StackedRNNCells`
* `ConvLSTM1D`, `ConvLSTM2D`, `ConvLSTM3D`

### Also supported

Expand All @@ -71,15 +81,11 @@ Would you like to build/train a model using Keras/Python? And would you like to
### Currently not supported are the following:

`Lambda` ([why](FAQ.md#why-are-lambda-layers-not-supported)),
`ConvLSTM1D`, `ConvLSTM2D`, `Discretization`,
`GRUCell`, `Hashing`,
`IntegerLookup`,
`LocallyConnected1D`, `LocallyConnected2D`,
`LSTMCell`, `Masking`,
`RepeatVector`, `RNN`, `SimpleRNN`,
`SimpleRNNCell`, `StackedRNNCells`, `StringLookup`, `TextVectorization`,
`Bidirectional`, `GRU`, `LSTM`, `CuDNNGRU`, `CuDNNLSTM`,
`ThresholdedReLU`, `temporal` models
`Hashing`, `HashedCrossing`,
`MelSpectrogram`, `STFTSpectrogram`,
`StringLookup`, `TextVectorization`,
stateful recurrent layers,
`temporal` models

Usage
-----
Expand Down
2 changes: 1 addition & 1 deletion include/fdeep/common.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,7 +80,7 @@ namespace internal {
using RowMajorMatrixXf = Eigen::Matrix<float_type, Eigen::Dynamic, Eigen::Dynamic, Eigen::RowMajor>;
using ArrayXf = Eigen::Array<float_type, Eigen::Dynamic, Eigen::Dynamic>;
using ArrayXf1D = Eigen::Array<float_type, Eigen::Dynamic, 1>;
using MappedRowMajorMatrixXf = Eigen::Map<RowMajorMatrixXf, Eigen::Aligned>;
using MappedRowMajorMatrixXf = Eigen::Map<const RowMajorMatrixXf, Eigen::Unaligned>;

inline float_type tanh_typed(float_type x)
{
Expand Down
Loading