GitLab

* upd

* fig edgebatch edges

* add test

* trigger

* Update README.md for pytorch PinSage example.

Add noting that the PinSage model example under
example/pytorch/recommendation only work with Python 3.6+
as its dataset loader depends on stanfordnlp package
which work only with Python 3.6+.

* Provid a frame agnostic API to test nn modules on both CPU and CUDA side.

1. make dgl.nn.xxx frame agnostic
2. make test.backend include dgl.nn modules
3. modify test_edge_softmax of test/mxnet/test_nn.py and
    test/pytorch/test_nn.py work on both CPU and GPU

* Fix style

* Delete unused code

* Make agnostic test only related to tests/backend

1. clear all agnostic related code in dgl.nn
2. make test_graph_conv agnostic to cpu/gpu

* Fix code style

* fix

* doc

* Make all test code under tests.mxnet/pytorch.test_nn.py
work on both CPU and GPU.

* Fix syntex

* Remove rand

* Add TAGCN nn.module and example

* Now tagcn can run on CPU.

* Add unitest for TGConv

* Fix style

* For pubmed dataset, using --lr=0.005 can achieve better acc

* Fix style

* Fix some descriptions

* trigger

* Fix doc

* Add nn.TGConv and example

* Fix bug

* Update data in mxnet.tagcn test acc.

* Fix some comments and code

* delete useless code

* Fix namming

* Fix bug

* Fix bug

* Add test for mxnet TAGCov

* Add test code for mxnet TAGCov

* Update some docs

* Fix some code

* Update docs dgl.nn.mxnet

* Update weight init

* Fix

* reproduce the bug

* Fix concurrency bug reported at #755.
Also make test_shared_mem_store.py more deterministic.

* Update test_shared_mem_store.py

* Update dmlc/core

* networkx >= 2.4 will break our examples

* Update tutorials/requirements

* fix selfloop edges

* upd version

* upd

* add test

* fix

* upd

* merge

* hotfix

* upd

* fix

* Add early stop

* add mxnet version

* Poke ci

* fix gat code to use latest edge softmax module

* avoid transpose

* update README

* use edge_softmax op

* mxnet edge softmax op

* mxnet gat

* update README

* fix unittest

* fix ci

* fix mxnet nn test; relax criteria for prod reducer

* Update gcn_mp.py

* Update train.py

* Update train.py

* Update train.py

* Update gat_batch.py

* Update gat_batch.py

* Update gcn_mp.py

* Update prerequisites of README

* dependencies for pytorch models

* dependencies for mxnet models

* minor

* two better GAT implementations

* update numbers

* use version switch for spmm

* add missing dropout and output heads

* mxgat

* pygat

* Update gat_batch.py

* add train

* Update gat_batch.py

* Update gat.py

* Update gat_batch.py

* readme

* Update and rename README to README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* fix gat.

* fix context.

* move gat to the new api.

* fix gcn.

* update sse.

* fix dgl core.

* update sse.

* fix small bugs in dgl core.

* fix mxnet tests.

* retrigger

* address comments and fix more bugs.

* fix

* fix tests.

* support mxnet.

* add mxnet version of GCN.

* rename mxnet.nd as F.

* add mxnet GAT.

* enable GPU for GCN.

* fix MXNet GCN train.

* Use adam to optimize GAT

* support more operators.

* support sparse arrays.

* update mxnet backend.

* support index_copy.

* remove NN.

* update mxnet backend.

* temp check in.

* fix data conversion.

* add test.

* clean up mxnet backend.

* update mxnet examples.

* Revert "remove NN."

This reverts commit d815d9a0ec619f9ce9099c48cd35db9d8e947483.

* temp disable MXNet version of NN.