17d604b5c7
[Feature] Allow using NCCL for communication in dgl.NodeEmbedding and dgl.SparseOptimizer ( #2824 )
...
* Split from NCCL PR
* Fix type in comment
* Expand documentation for sparse_all_to_all_push
* Restore previous behavior in example
* Re-work optimizer to use NCCL based on gradient location
* Allow for running with embedding on CPU but using NCCL for gradient exchange
* Optimize single partition case
* Fix pylint errors
* Add missing include
* fix gradient indexing
* Fix line continuation
* Migrate 'first_step'
* Skip tests without enough GPUs to run NCCL
* Improve empty tensor handling for pytorch 1.5
* Fix indentation
* Allow multiple NCCL communicator to coexist
* Improve handling of empty message
* Update python/dgl/nn/pytorch/sparse_emb.py
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com >
* Update python/dgl/nn/pytorch/sparse_emb.py
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com >
* Keepy empty tensor dimensionaless
* th.empty -> th.tensor
* Preserve shape for empty non-zero dimension tensors
* Use shared state, when embedding is shared
* Add support for gathering an embedding
* Fix typo
* Fix more typos
* Fix backend call
* Use NodeDataLoader to take advantage of ddp
* Update training script to share memory
* Only squeeze last dimension
* Better handle empty message
* Keep embedding on the target device GPU if dgl_sparse if false in RGCN example
* Fix typo in comment
* Add asserts
* Improve documentation in example
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com >
2021-06-10 21:19:00 -07:00
0e964ea566
increase timeout for dgl.nn.NodeEmbedding TCPStore get/wait to 10 minutes ( #2967 )
...
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com >
2021-06-02 14:56:42 +08:00
fcfe52aee5
fix residual ( #2962 )
2021-06-01 13:50:39 +08:00
5d7e80f413
[Feature] Add bruteforce implementation for KNN with O(Nk) space complexity ( #2892 )
...
* add bruteforce impl
* add support for bruteforce-sharemem
* modify python API
* add tests
* change file path
* change python API
* fix lint
* fix test
* also check worst_dist in the last few dim
* use heap and early-stop on CPU
* fix lint
* fix lint
* add device check
* use cuda function to determine max shared mem
* use cuda to determine block info
* add memory free for tmp var
* update doc-string and add dist option
* fix lint
* add more tests
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com >
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com >
2021-05-19 18:27:43 +08:00
50492d56a8
Fixing issue #2921 ( #2922 )
...
* fixing #2921
* fixing #2921
2021-05-18 09:54:38 +08:00
b03077b60d
[NN] GCNII model added < https://arxiv.org/pdf/2007.02133.pdf > ( #2874 )
...
* Auto stash before rebase of "origin/pytorch-nn-working"
GCNII model added
* linting
* linting
* lint
* Frequency Adaptive gcn init comit
* Revert "Frequency Adaptive gcn init comit"
This reverts commit 86a80586ac0040497c1edfa0e80df719992dcc4a.
* Update python/dgl/nn/pytorch/conv/gcn2conv.py
modified docstring
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com >
* added beta formula and changed param name
* fix docstring
* lint
* white space lint
* update docstring
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com >
* docstring formula update
* added gcn2
* added GCN2Conv
* Update nn.pytorch.rst
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com >
2021-05-17 10:25:59 +08:00
9983107384
[Feature] Support multidimensional features for GAT ( #2912 )
...
* support multidimensional features for GAT
* docstring
* lint
* fix
2021-05-15 02:47:23 +08:00
c18f957d0e
[Model] add implementation of twirls ( #2770 )
...
* add implementation of twirls
* format the code
* fix some format error, and ignore others
* fix format errors
* fix format errors
* expose unfolding & attention
* Update nn.pytorch.rst
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com >
2021-05-11 09:58:48 +08:00
e18c2ab408
[NN] Fix GATConv for Broadcasting with Residual Connections ( #2867 )
...
* Update
* update
Co-authored-by: Ubuntu <ubuntu@ip-172-31-59-108.us-west-2.compute.internal >
2021-04-27 14:36:06 +08:00
c37e036432
[Bug Fix] Fix sparse opt bug ( #2859 )
...
* Fix #2856
* upd
* Fix unitest
* upd
* upd
* upd
* Fix
Co-authored-by: Ubuntu <ubuntu@ip-172-31-57-25.ec2.internal >
2021-04-25 13:40:21 +08:00
e83d0a8007
[Feature] Add kd-tree implementation (CPU) for kNN ( #2767 )
...
* add submodule nanoflann
* finish python API for knn
* finish ndarray adaptor
* finish cpu-kdtree version of knn
* use openmp
* add endline
* upt
* upt
* fix format and code style
* upt
* add warning for gpu-cpu copy
* avoid contiguous copy
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com >
Co-authored-by: Tong He <hetong007@gmail.com >
2021-04-09 17:39:34 +08:00
cfe6e70b7a
RGAT baseline for OGB-LSC node classification challenge ( #2810 )
...
* RGAT baseline
* update links (data.dgl.ai hasn't refreshed yet)
* nvm still use dgl.ai domain
* descriptions
* actually it took less
* address comments
* stop worrying about cache
* update links
* oops
2021-04-05 18:21:18 +08:00
03482f0a4c
[NN] tf nn for edgeConv ( #2741 )
...
* tf nn for edgeConv
* Auto stash before merge of "tf_working" and "origin/tf_working"
* clean up
* added test for edge_conv
* fix
* fix
* fix
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com >
Co-authored-by: Quan Gan <coin2028@hotmail.com >
2021-03-25 13:44:07 +08:00
edf644633d
[performance] Optimize the association order of AXW in GraphSAGE. ( #2747 )
...
* upd
* lint
* upd
* upd
* compatibility
* upd
* upd
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com >
2021-03-19 01:41:33 +08:00
366cc7ebf3
[Pickle] Fix HeteroGraphConv pickle problem ( #2761 )
...
* fix pickle problem
* lint
* add pickle tests
* fix
* fix
* fix
* fix
* fix for windows
2021-03-18 17:56:28 +08:00
09ef2c2c3d
[Doc] Add Pseudo Code for HeteroGraphConv ( #2729 )
...
* Update
* Update
* Update
* Update
Co-authored-by: Ubuntu <ubuntu@ip-172-31-15-8.us-west-2.compute.internal >
2021-03-17 20:00:15 +08:00
3c8382f6b6
Fix typos in zero-in-degree ( #2750 )
2021-03-14 19:13:59 +08:00
7380d61eb9
[NN] Support Unidirectional Bipartite Graphs in CFConv ( #2674 )
...
* Update
* update
* Update
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com >
2021-03-03 16:35:07 +08:00
71d53d9eef
add thumbnails for tutorials ( #2695 )
...
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com >
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com >
2021-02-23 20:33:35 +08:00
174dd18905
[hotfix] A bunch of fix for SetTransformer ( #2658 )
...
* upd
* lint
* upd
* fix
* fix
* upd
* upd
* fix
* warning
* upd
* upd
2021-02-23 14:19:51 +08:00
453d358d2c
[Doc] Fix docs for sparse optimizer ( #2680 )
...
Co-authored-by: Ubuntu <ubuntu@ip-172-31-56-220.ec2.internal >
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com >
2021-02-19 14:07:35 +08:00
dda103d9f6
Fix bug in EdgeConv implementation ( #2669 )
...
* Update edgeconv.py
* update edgeconv to builtin
* add explanation on why ReLU is not there
* remove redundant code
2021-02-19 13:07:21 +08:00
0cbaa1fbcb
[Doc] Make the docstring of nn.glob more informative. ( #2649 )
...
* upd
* fix weight and sum
* upd sortpool
2021-02-12 12:29:22 +08:00
9e630101a1
fix and lots of tests ( #2650 )
2021-02-12 11:40:05 +08:00
3ca52b1e59
[Fix] Various Fix after Bug Bash ( #2620 )
...
* update
* Misc fix
* Update
2021-02-05 17:34:16 +08:00
23afe911b2
[Bug Fix] Fix several sparse optimizer bugs ( #2596 )
...
* Fix pytorch TCP kvstore bug
* lint
* Fix
* upd
* Fix lint
* Fix
* trigger
* fix
Co-authored-by: Ubuntu <ubuntu@ip-172-31-56-220.ec2.internal >
2021-02-05 11:49:41 +08:00
1f6eba9ed3
[Bug] Heterogeneous graph convolution bugfix ( #2578 )
...
* fix heterograph conv
* remove test cases
* fix test
* fix test
* fix test
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com >
2021-01-29 09:30:19 +08:00
e4ddafe951
[NN] add multihead in DotGatConv ( #2549 )
...
* add multihead in DotGatConv
* Fix spacing issue
* Add Unit test for dotgat
* Modified Unit test for dotgat
* Add transformer like divisor
* Update dotgatconv.py
Co-authored-by: Chen <chesirui@3c22fbe5458c.ant.amazon.com >
Co-authored-by: Zihao Ye <expye@outlook.com >
2021-01-28 21:05:10 +08:00
a7e941c379
[Feature] Add support for sparse embedding ( #2451 )
...
* Add sparse embedding for dgl and update rgcn example
* upd
* Fix
* Revert "Fix"
This reverts commit 4da87cdfb8b8c3506b7fc7376cd2385ba8045c2a.
* Fix
* upd
* upd
* Fix
* Add unitest and update impl
* fix
* Clean up rgcn example code
* upd
* upd
* update
* Fix
* update score
* sparse for sage
* remove model sparse
* upd
* upd
* remove global norm
* revert delete model_sparse.py
* update according to comments
* Fix doc
* upd
* Fix test
* upd
* lint
* lint
* lint
* upd
* upd
* clean up
Co-authored-by: Ubuntu <ubuntu@ip-172-31-56-220.ec2.internal >
2021-01-28 00:26:49 +08:00
0855d25501
[NN] Support scalar edge weight for GraphConv, SAGEConv and GINConv ( #2557 )
...
* add edge weight in forward
* fix lint
* fix
* fix
* address comments
* add utils
* add util to normalize in gcn way
* fix lint
* add unittest
* fix lint
* fix docstring
* fix docstring
* address comments
* improve notation consistence
* use preferred fn
2021-01-26 23:34:12 +08:00
6c23fba8fa
[Doc][Org] move edge_softmax to functional ( #2442 )
...
* move edge_softmax to functional
* change examples
* fixes
* revert
* fix
* fix
* remove
* fix
* test
* add init
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com >
2021-01-26 12:46:37 +08:00
bedccc2392
fix 2532 ( #2533 )
2021-01-14 13:40:25 +08:00
54f186bba0
Fix ( #2524 )
...
Co-authored-by: Ubuntu <ubuntu@ip-172-31-59-204.ec2.internal >
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com >
2021-01-13 15:09:21 +08:00
498505bafb
[Doc] fix docserver build for tutorials ( #2518 )
...
* fix doc build
* hide GPU code
2021-01-12 17:55:58 +08:00
d3c1645550
[Performance][Test] New low memory implementation for RGCN and related regression tests ( #2468 )
...
* WIP
* finish lowmem impl; benchmarking
* wip
* wip
* fix benchmarks
* fix bug in searchsorted
* update readme
* update
* lint and ogb dependency
* add flags to to_homogeneous
* fix docstring
* address comments
2021-01-04 22:23:59 +08:00
baace710cf
[documentation] Improve the docstring of global pooling layer ( #2437 )
...
* upd
* fix-link
* lint
2020-12-21 10:12:54 +08:00
492ad9be81
[Fix] Avoid Overloading Nonzero for PyTorch Backend ( #2434 )
...
* Update gatedgraphconv.py
* Update entity_classify.py
* Update data-process.rst
* Update reading_data.py
* Update data-process.rst
* Update utils.py
* Update knowledge_graph.py
* Update entity_classify.py
* Update rdf.py
* Update entity_classify_mb.py
* Update test_classify.py
* Update tensor.py
* Update sparse.py
* Update entity_classify_mp.py
* Update 6_line_graph.py
2020-12-20 13:12:17 +08:00
013d145663
[NN] Attention Retrieval for NN Modules ( #2397 )
...
* Update
* Update
2020-12-07 23:42:25 +08:00
45e3e9af2c
[WIP] Misc Fix ( #2359 )
...
* Update
* Update
* Update
2020-11-19 23:54:55 +08:00
812f597096
Update appnpconv.py ( #2324 )
2020-11-10 15:37:30 +08:00
6b0d42db15
Fix typo ( #2235 )
...
Co-authored-by: Chao Ma <mctt90@gmail.com >
2020-09-25 16:40:22 +08:00
0435b74c48
[Hotfix] Fix layer norm ( #2119 )
...
* hotfix
* Fix Layer Norm
Co-authored-by: Ubuntu <ubuntu@ip-172-31-87-240.ec2.internal >
2020-08-28 11:20:33 +08:00
7ead28c315
[Doc] [NN] Fix message and nn user guide link ( #2090 )
...
* fix user guide link
* some fix on tensorflow
2020-08-22 00:18:22 +08:00
f13b9b6212
[Doc] Scan the API docs and make many changes ( #2080 )
...
* WIP: api
* dgl.sampling, dgl.data
* dgl.sampling; dgl.dataloading
* sampling packages
* convert
* subgraph
* deprecate
* subgraph APIs
* All docstrings for convert/subgraph/transform
* almost all funcs under dgl namespace
* WIP: DGLGraph
* done graph query
* message passing functions
* lint
* fix merge error
* fix test
* lint
* fix
Co-authored-by: Quan Gan <coin2028@hotmail.com >
2020-08-20 23:26:49 +08:00
65866989c0
[NN] [Doc] Fix nn api doc ( #2047 )
...
* fix mxnet nn doc and no note on chebconv
* change notes to note to highlight
2020-08-18 23:47:16 +08:00
be444e52d9
[Doc/Feature] Refactor, doc update and behavior fix for graphs ( #1983 )
...
* Update graph
* Fix for dgl.graph
* from_scipy
* Replace canonical_etypes with relations
* from_networkx
* Update for hetero_from_relations
* Roll back the change of canonical_etypes to relations
* heterograph
* bipartite
* Update doc
* Fix lint
* Fix lint
* Fix test cases
* Fix
* Fix
* Fix
* Fix
* Fix
* Fix
* Update
* Fix test
* Fix
* Update
* Use DGLError
* Update
* Update
* Update
* Update
* Fix
* Fix
* Fix
* Fix
* Fix
* Fix
* Fix
* Fix
* Update
* Fix
* Update
* Fix
* Fix
* Fix
* Update
* Fix
* Update
* Fix
* Update
* Update
* Update
* Update
* Update
* Update
* Update
* Fix
* Fix
* Update
* Update
* Update
* Update
* Update
* Update
* rewrite sanity checks
* delete unnecessary checks
* Update
* Update
* Update
* Update
* Update
* Update
* Update
* Update
* Fix
* Update
* Update
* Update
* Fix
* Fix
* Fix
* Update
* Fix
* Update
* Fix
* Fix
* Update
* Fix
* Update
* Fix
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com >
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com >
Co-authored-by: Quan Gan <coin2028@hotmail.com >
2020-08-18 04:26:29 +08:00
0afc3cf8e1
[NN] Add ChebConv for Tensorflow ( #2038 )
...
* add chebconv
* fix
* lint
* fix lint
* docs
2020-08-17 22:12:27 +08:00
b1e69105d4
[Doc] API Doc update for mxnet and tf, remove some degree check ( #2028 )
...
* mx tf relconv
* use method instead of private attr
* src and dst have different fc for gat
* update edgeconv
* change sage and sgconv
* no degree check on gin
* add remainding API doc
* fix pylint
* infer fc_src and fc_dst, only one tensor for block
* fix pytest
2020-08-17 13:23:44 +08:00
2fa7f71ad5
[Doc] new nn api doc ( #2019 )
...
* Add dotproduct attention
* [Feature] Add dotproduct attention
* [Feature] Add dotproduct attention
* [Feature] Add dotproduct attention
* [New] Update landing page
* [New] Update landing page
* [New] Update landing page
* [Doc] Update landing page
* [Doc] Update landing page
* [Doc] Update landing page
* [Doc] Update landing page
* [Doc] Update landing page
* [Doc] Update landing page
* [Doc] Update landing page
* [Doc] Update landing page
* [Improvement] use dgl build-in in dotgatconv
* [Doc] review API doc string bottom up
* [Doc] Add doc of input and output features
* [Doc] Update doc string for pooling and transformer.
* [Doc] Reformat doc string and change some wordings.
* [Doc] Doc string refactoring.
* [Doc] Doc string refactoring.
* [Doc] Doc string refactoring.
* [Doc] Doc string refactoring.
* [Doc] Doc string refactoring.
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com >
2020-08-14 18:00:38 +08:00
4672c022a8
[hotfix] no in-degree check on sage ( #2014 )
...
* no degre check on sage
* fix pytest
* no in degree check for heteroconv
2020-08-13 13:11:55 +08:00
nv-dlasalle