Layers cheatsheet

docs/source/api_reference/keras_layers.csv

@@ -0,0 +1,36 @@
+Layer,Reference, Model
+InteratomicL2Distances,,
+GraphConv,`ref <https://arxiv.org/abs/1509.09292>`_,
+GraphPool, `ref <https://arxiv.org/abs/1509.09292>`_,
+GraphGather, `ref <https://arxiv.org/abs/1509.09292>`_,
+MolGANConvolutionLayer,`ref <https://arxiv.org/abs/1805.11973>`_, MolGan
+MolGANAggregationLayer, `ref <https://arxiv.org/abs/1805.11973>`_, MolGan
+MolGANMultiConvolutionLayer, `ref <https://arxiv.org/abs/1805.11973>`_, MolGan
+MolGANEncoderLayer, `ref <https://arxiv.org/abs/1805.11973>`_, MolGan
+LSTMStep,,
+AttnLSTMEmbedding, `ref <https://arxiv.org/abs/1606.04080>`_,
+IterRefLSTMEmbedding, ,
+SwitchedDropout, ,
+WeightedLinearCombo, ,
+CombineMeanSt, ,
+Stack,,
+VinaFreeEnergy,,
+NeighborList,,
+AtomicConvolution, `ref <https://arxiv.org/abs/1703.10603>`_,
+AlphaShareLayer,, Sluice Network
+SluiceLoss,, Sluice Network
+BetaShare,, Sluice Network
+ANIFeat,,
+GraphEmbedPoolLayer,`ref <https://arxiv.org/abs/1703.00792>`_,
+Highway,`ref <https://arxiv.org/abs/11505.00387>`_,
+WeaveLayer,`ref <https://pubmed.ncbi.nlm.nih.gov/27558503/>`_,
+WeaveGather,`ref <https://pubmed.ncbi.nlm.nih.gov/27558503/>`_,
+DTNNEmbedding, ,
+DTNNStep, ,
+DTNNGather,,
+DAGLayer, `ref <https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3739985/>`_,
+DAGGather,,
+MessagePassing, `ref <https://arxiv.org/abs/1511.06391>`_,
+EdgeNetwork, `ref <https://arxiv.org/abs/1511.06391>`_, MessagePassing
+GatedRecurrentUnit, `ref <https://arxiv.org/abs/1511.06391>`_, MessagePassing
+SetGather,,

docs/source/api_reference/layers.rst

@@ -8,6 +8,8 @@ transformations. For now, most layers are Keras only but over
 time we expect this support to expand to other types of models
 and layers.
 
+.. include:: layers_cheatsheet.rst
+
 Keras Layers
 ------------
 

docs/source/api_reference/layers_cheatsheet.rst

@@ -0,0 +1,29 @@
+Layers Cheatsheet
+-----------------
+The "layers cheatsheet" lists various scientifically relevant differentiable layers implemented in DeepChem.
+
+Note that some layers implemented for specific model architectures such as :code:`GROVER`
+and :code:`Attention` layers, this is indicated in the `Model` column of the table.
+
+In order to use the layers, make sure that the backend (Keras and tensorflow, Pytorch or Jax) is installed.
+
+**Tensorflow Keras Layers**
+
+These layers are subclasses of the :code:`tensorflow.keras.layers.Layer` class.
+
+.. csv-table:: Custom Keras Layers
+    :file: ./keras_layers.csv
+    :width: 100%
+    :header-rows: 1
+
+**PyTorch**
+
+These layers are subclasses of the :code:`torch.nn.Module` class.
+
+.. csv-table:: Custom PyTorch Layers
+    :file: ./torch_layers.csv
+    :width: 100%
+    :header-rows: 1
+
+
+

docs/source/api_reference/torch_layers.csv

@@ -0,0 +1,24 @@
+Layer,Reference, Model
+MultilayerPerceptron,,
+ScaleNorm,`ref <https://arxiv.org/abs/2002.08264>`_, Molecular Attention Transformer
+MATEncoderLayer,`ref <https://arxiv.org/abs/2002.08264>`_, Molecular Attention Transformer
+MultiHeadedMATAttention,`ref <https://arxiv.org/abs/2002.08264>`_, Molecular Attention Transformer
+SublayerConnection,`ref <https://arxiv.org/abs/1706.03762>`_, Transformer
+MATEmbedding,`ref <https://arxiv.org/abs/2002.08264>`_, Molecular Attention Transformer
+MATGenerator, `ref <https://arxiv.org/abs/2002.08264>`_, Molecular Attention Transformer
+Affine,`ref <http://arxiv.org/abs/2110.15828>`_, Normalizing Flow
+RealNVPLayer,`ref <http://arxiv.org/abs/2110.15828>`_, Normalizing Flow
+DMPNNEncoderLayer,`ref <https://arxiv.org/pdf/1904.01561.pdf>`_, Normalizing Flow
+PositionwiseFeedForward,`ref <https://arxiv.org/abs/2002.08264>`_, Molecular Attention Transformer
+GraphPool, `ref <https://arxiv.org/abs/1509.09292>`_,
+GroverMPNEncoder, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+GroverAttentionHead, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+GroverMTBlock, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+GroverTransEncoder, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+GroverEmbedding, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+GroverAtomVocabPredictor, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+GroverBondVocabPredictor, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+GroverFunctionalGroupPredictor, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover
+ScaledDotProductAttention, `ref <https://arxiv.org/abs/1706.03762>`_, Transformer
+SelfAttention, `ref <https://arxiv.org/abs/1706.03762>`_, Transformer
+GroverReadout, `ref <https://drug.ai.tencent.com/publications/GROVER.pdf>`_, Grover