[Docs] Adjust doc (#808)

* fix doc

* poke ci

* poke ci

* upd

* fix doc

* upd

* add title for all nn

* upd

.github/PULL_REQUEST_TEMPLATE.md

@@ -4,7 +4,7 @@ It'll be great if relevant reviewers can be assigned as well.-->
 
 ## Checklist
 Please feel free to remove inapplicable items for your PR.
-- [ ] The PR title starts with [$CATEGORY] (such as [Model], [Doc], [Feature]])
+- [ ] The PR title starts with [$CATEGORY] (such as [NN], [Model], [Doc], [Feature]])
 - [ ] Changes are complete (i.e. I finished coding on this PR)
 - [ ] All changes have test coverage
 - [ ] Code is well-documented

docs/source/api/python/index.rst

@@ -18,3 +18,4 @@ API Reference
    subgraph
    graph_store
    nodeflow
+   random

docs/source/api/python/nn.mxnet.rst

 .. _apinn-mxnet:
 
-dgl.nn.mxnet
-============
+NN Modules (MXNet)
+===================
 
-dgl.nn.mxnet.conv
------------------
+.. contents:: Contents
+    :local:
+
+We welcome your contribution! If you want a model to be implemented in DGL as a NN module,
+please `create an issue <https://github.com/dmlc/dgl/issues>`_ started with "[Feature Request] NN Module XXXModel".
+
+If you want to contribute a NN module, please `create a pull request <https://github.com/dmlc/dgl/pulls>`_ started
+with "[NN] XXXModel in MXNet NN Modules" and our team member would review this PR.
+
+Conv Layers 
+----------------------------------------
 
 .. automodule:: dgl.nn.mxnet.conv
 
+GraphConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.mxnet.conv.GraphConv
     :members: weight, bias, forward
     :show-inheritance:
 
+RelGraphConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.mxnet.conv.RelGraphConv
     :members: forward
     :show-inheritance:
 
+TAGConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.mxnet.conv.TAGConv
     :members: forward
     :show-inheritance:
 
-dgl.nn.mxnet.glob
------------------
+Global Pooling Layers 
+----------------------------------------
 
 .. automodule:: dgl.nn.mxnet.glob
+
+SumPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: dgl.nn.mxnet.glob.SumPooling
+    :members:
+    :show-inheritance:
+
+AvgPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: dgl.nn.mxnet.glob.AvgPooling
+    :members:
+    :show-inheritance:
+
+MaxPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: dgl.nn.mxnet.glob.MaxPooling
+    :members:
+    :show-inheritance:
+
+SortPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: dgl.nn.mxnet.glob.SortPooling
     :members:
     :show-inheritance:
 
-dgl.nn.mxnet.softmax
---------------------
+GlobalAttentionPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: dgl.nn.mxnet.glob.GlobalAttentionPooling
+    :members:
+    :show-inheritance:
+
+Set2Set
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: dgl.nn.mxnet.glob.Set2Set
+    :members:
+    :show-inheritance:
+
+
+Utility Modules
+----------------------------------------
+
+Edge Softmax
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. automodule:: dgl.nn.mxnet.softmax
     :members: edge_softmax

docs/source/api/python/nn.pytorch.rst

 .. _apinn-pytorch:
 
-dgl.nn.pytorch
-==============
+NN Modules (PyTorch)
+====================
 
-dgl.nn.pytorch.conv
--------------------
+.. contents:: Contents
+    :local:
+
+We welcome your contribution! If you want a model to be implemented in DGL as a NN module,
+please `create an issue <https://github.com/dmlc/dgl/issues>`_ started with "[Feature Request] NN Module XXXModel".
+
+If you want to contribute a NN module, please `create a pull request <https://github.com/dmlc/dgl/pulls>`_ started
+with "[NN] XXXModel in PyTorch NN Modules" and our team member would review this PR.
+
+Conv Layers 
+----------------------------------------
 
 .. automodule:: dgl.nn.pytorch.conv
 
+GraphConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.GraphConv
     :members: weight, bias, forward, reset_parameters
     :show-inheritance:
     
+RelGraphConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.RelGraphConv
     :members: forward
     :show-inheritance:
     
+TAGConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.TAGConv
     :members: forward
     :show-inheritance:
     
+GATConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.GATConv
     :members: forward
     :show-inheritance:
     
+EdgeConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.EdgeConv
     :members: forward
     :show-inheritance:
     
+SAGEConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.SAGEConv
     :members: forward
     :show-inheritance:
     
+SGConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.SGConv
     :members: forward
     :show-inheritance:
     
+APPNPConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.APPNPConv
     :members: forward
     :show-inheritance:
     
+GINConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.GINConv
     :members: forward
     :show-inheritance:
     
+GatedGraphConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.GatedGraphConv
     :members: forward
     :show-inheritance:
     
+GMMConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.GMMConv
     :members: forward
     :show-inheritance:
     
+ChebConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.ChebConv
     :members: forward
     :show-inheritance:
     
+AGNNConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.AGNNConv
     :members: forward
     :show-inheritance:
     
+NNConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.NNConv
     :members: forward
     :show-inheritance:
     
+Dense Conv Layers
+----------------------------------------
+
+DenseGraphConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.DenseGraphConv
     :members: forward
     :show-inheritance:
     
+DenseSAGEConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.DenseSAGEConv
     :members: forward
     :show-inheritance:
     
+DenseChebConv
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.conv.DenseChebConv
     :members: forward
     :show-inheritance:
     
-dgl.nn.pytorch.glob
--------------------
+Global Pooling Layers 
+----------------------------------------
+
 .. automodule:: dgl.nn.pytorch.glob
 
+SumPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.SumPooling
     :members:
     :show-inheritance:
     
+AvgPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.AvgPooling
     :members:
     :show-inheritance:
     
+MaxPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.MaxPooling
     :members:
     :show-inheritance:
     
+SortPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.SortPooling
     :members:
     :show-inheritance:
     
+GlobalAttentionPooling
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.GlobalAttentionPooling
     :members:
     :show-inheritance:
     
+Set2Set
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.Set2Set
     :members: forward
     :show-inheritance:
     
+SetTransformerEncoder
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.SetTransformerEncoder
     :members:
     :show-inheritance:
     
+SetTransformerDecoder
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.glob.SetTransformerDecoder
     :members:
     :show-inheritance:
     
-dgl.nn.pytorch.softmax
-----------------------
+Utility Modules
+----------------------------------------
 
-.. automodule:: dgl.nn.pytorch.softmax
-    :members: edge_softmax
-
-dgl.nn.pytorch.factory
-----------------------
-
-.. automodule:: dgl.nn.pytorch.conv
+KNNGraph
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. autoclass:: dgl.nn.pytorch.factory.KNNGraph
     :members:
     :show-inheritance:
     
+SegmentedKNNGraph
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
 .. autoclass:: dgl.nn.pytorch.factory.SegmentedKNNGraph
     :members:
     :show-inheritance:
+    
+Edge Softmax
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. automodule:: dgl.nn.pytorch.softmax
+    :members: edge_softmax
+    
+

docs/source/api/python/nn.rst

 .. _apinn:
 
-dgl.nn
-======
+NN Modules
+==========
 
 .. automodule:: dgl.nn
 

docs/source/contribute.rst

@@ -10,6 +10,7 @@ General development process
 A non-inclusive list of types of contribution is as follows:
 
 * New features and enhancements (`example <https://github.com/dmlc/dgl/pull/331>`__).
+* New NN Modules (`example <https://github.com/dmlc/dgl/pull/788>`__).
 * Bugfix (`example <https://github.com/dmlc/dgl/pull/247>`__).
 * Document improvement (`example <https://github.com/dmlc/dgl/pull/263>`__).
 * New models and examples (`example <https://github.com/dmlc/dgl/pull/279>`__).

docs/source/features/nn.rst

+.. currentmodule:: dgl
+
+NN Modules
+===============
+
+A set of high-level pre-defined modules are provided to build graph neural networks.
+
+
+.. toctree::
+   :glob:
+   :caption: NN Modules
+ 
+   ../api/python/nn
+
+
+Contribution Guide
+~~~~~~~~~~~~~~~~~~~~~
+
+We welcome your contribution! If you want a model to be implemented in DGL as a NN module,
+please `create an issue <https://github.com/dmlc/dgl/issues>`_ started with "[Feature Request] NN Module XXXModel".
+
+If you want to contribute a NN module, please `create a pull request <https://github.com/dmlc/dgl/pulls>`_ started
+with "[NN] XXXModel in MXNet/PyTorch NN Modules" and our team member would review this PR.

docs/source/index.rst

@@ -170,6 +170,7 @@ Or go through all of them :doc:`here <tutorials/models/index>`.
    :glob:
 
    features/builtin
+   features/nn
 
 .. toctree::
    :maxdepth: 3