文件历史

29 次代码提交

作者 SHA1 备注 提交日期
peizhou001 3a2a5031e9 [API Deprecation]Deprecate candidates in dgl.distributed (#5116) 2023-01-06 16:17:44 +08:00
Rhett Ying cd3fa03035 [Dist] init node/edge data store for Node/EdgeDataView in appropriate place (#4906)
* [Dist] instantiate NodeDataView in lazy mode

* fix test failure

* init node/edge data store at the very beginning

* fix test failures

* refine comment

* add more tests
2022-11-21 09:52:36 +08:00
Rhett Ying 6e1cc7dac2 [Dist] fix is_node check according to policy (#4911)
* [Dist] fix is_node check according to policy

* add more tests
2022-11-17 09:27:50 +08:00
Rhett Ying 1e8c645d8a [Dist] fix max length to 80 introduced in deprecate etype PR (#4809) 2022-11-04 11:48:36 +08:00
Rhett Ying ed8e9c4440 [Dist] deprecate etype and always use canonical etype for partition and load (#4777)
* [Dist] deprecate etype and always use canonical etype for partition and load

* enable canonical etypes in dist part pipeline

* resolve rebase conflicts

* fix lint

* fix test failure

* throw exception if outdated part config is loaded

* refine

* refine

* revert unnecessary change

* fix typo
2022-11-04 10:08:18 +08:00
Quan (Andy) Gan 2bca4759c1 [Sampling] Enable sampling with edge masks in sample_etype_neighbors (#4749)
* sample neighbors with masks

* oops

* refactor again

* remove

* remove debug code

* rename macro

* address comments

* more stuff

* remove

* fix

* try fix unit test

* oops

* fix test

* oops

* change name

* rename a lot of stuff

* oops

* ugh

* misc fixes

* lint

* address a lot of comments

* lint

* lint

* fix

* that was silly

* fix

* fix

* fix

* oops
2022-10-29 15:49:44 +08:00
Hongzhi (Steve), Chen a208e8868b [Misc] Black auto fix. (#4680)
* [Misc] Black auto fix.

* fix pylint disable

Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>
2022-10-08 14:15:56 +08:00
Rhett Ying 39987bc599 [Feature] enable graph partition book support canonical etypes (#4343)
* [Feature] enable graph partition book support canonical etypes

* fix lint

* fix lint

* add todo

* refine according to review comments

* fix lint

* refine naming

* revert PartitionPolicy __init__

* refine docstring

* fix doc string
2022-08-16 11:52:02 +08:00
Rhett Ying 53835bdb75 [Dist][BugFix] enable sampling on bipartite (#4014)
* [Dist][BugFix] enable sampling on bipartite

* add comments for tests
2022-05-18 10:14:54 +08:00
Quan (Andy) Gan 92db4bd5a0 fix (#3700) 2022-01-28 20:51:15 +08:00
Quan (Andy) Gan da877df238 [CI] Migrate to pylint 2.6.0 (#3698)
* migrate to pylint 2.6.0

* fix

* fix?

* ???

* oops
2022-01-28 18:32:18 +08:00
nv-dlasalle 7f9279397a [Feature][DistDGL] Add NCCL support for range based partitions (#3213)
* Implement range based NDArrayPartition

* Finish implement range based partition support

* Add unit test

* Fix whitepace

* Add Kernel suffix

* Fix argument passing

* Add doxygen docs and improve variable naming

* Add unit test

* Add function for converting a partition book

* Add example to partition_op docs

* Fix dtype conversion for mxnet and tensorflow
2021-08-19 21:17:55 -07:00
Da Zheng 34426a98e9 [Distributed] Distributed heterograph training (#3069)
* support hetero RGCN.

* fix.

* simplify code.

* sample_neighbors return heterograph directly.

* avoid using to_heterogeneous.

* compute canonical etypes in advance.

* fix tests.

* fix.

* fix distributed data loader for heterograph.

* use NodeDataLoader.

* fix bugs in partitioning on heterogeneous graphs.

* fix lint.

* fix tests.

* fix.

* fix.

* fix bugs.

* fix tests.

* fix.

* enable coo for distributed.

* fix.

* fix.

* fix.

* fix.

* fix.

Co-authored-by: Ubuntu <ubuntu@ip-172-31-71-112.ec2.internal>
Co-authored-by: Zheng <dzzhen@3c22fba32af5.ant.amazon.com>
2021-07-17 10:57:00 +08:00
Da Zheng 0fc649523b [Doc] Distributed heterogeneous graph training (#2654)
* update DistGraph docstrings.

* add user guide.

* add doc string.

* fix.

* fix.

* fix.

Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
2021-02-19 10:48:05 -08:00
Da Zheng 25ac334403 [Distributed] Heterogeneous graph support (#2457)
* Distributed heterograph (#3)

* heterogeneous graph partition.

* fix graph partition book for heterograph.

* load heterograph partitions.

* update DistGraphServer to support heterograph.

* make DistGraph runnable for heterograph.

* partition a graph and store parts with homogeneous graph structure.

* update DistGraph server&client to use homogeneous graph.

* shuffle node Ids based on node types.

* load mag in heterograph.

* fix per-node-type mapping.

* balance node types.

* fix for homogeneous graph

* store etype for now.

* fix data name.

* fix a bug in example.

* add profiler in rgcn.

* heterogeneous RGCN.

* map homogeneous node ids to hetero node ids.

* fix graph partition book.

* fix DistGraph.

* shuffle eids.

* verify eids and their mappings when loading a partition.

* Id map from homogneous Ids to per-type Ids.

* verify partitioned results.

* add test for distributed sampler.

* add mapping from per-type Ids to homogeneous Ids.

* update example.

* fix DistGraph.

* Revert "add profiler in rgcn."

This reverts commit 36daaed8b660933dac8f61a39faec3da2467d676.

* add tests for homogeneous graphs.

* fix a bug.

* fix test.

* fix for one partition.

* fix for standalone training and evaluation.

* small fix.

* fix two bugs.

* initialize projection matrix.

* small fix on RGCN.

* Fix rgcn performance (#17)

Co-authored-by: Ubuntu <ubuntu@ip-172-31-62-171.ec2.internal>

* fix lint.

* fix lint.

* fix lint.

* fix lint.

* fix lint.

* fix lint.

* fix.

* fix test.

* fix lint.

* test partitions.

* remove redundant test for partitioning.

* remove commented code.

* fix partition.

* fix tests.

* fix RGCN.

* fix test.

* fix test.

* fix test.

* fix.

* fix a bug.

* update dmlc-core.

* fix.

* fix rgcn.

* update readme.

* add comments.

Co-authored-by: Ubuntu <ubuntu@ip-172-31-2-202.us-west-1.compute.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-132.us-west-1.compute.internal>
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-62-171.ec2.internal>

* fix.

* fix.

* add div_int.

* fix.

* fix.

* fix lint.

* fix.

* fix.

* fix.

* adjust.

* move code.

* handle heterograph.

* return pytorch tensor in GPB.

* remove some tests in example.

* add to_block for distributed training.

* use distributed to_block.

* remove unnecessary function in DistGraph.

* remove distributed to_block.

* use pytorch tensor.

* fix a bug in ntypes and etypes.

* enable norm.

* make the data loader compatible with the old format.

* fix.

* add comments.

* fix a bug.

* add test for heterograph.

* support partition without reshuffle.

* add test.

* support partition without reshuffle.

* fix.

* add test.

* fix bugs.

* fix lint.

* fix dataset.

* fix for mxnet.

* update docstring.

* rename to floor_div

* avoid exposing NodePartitionPolicy and EdgePartitionPolicy.

* fix docstring.

* fix error.

* fixes.

* fix comments.

* rename.

* rename.

* explain IdMap.

* fix docstring.

* fix docstring.

* update docstring.

* remove the code of returning heterograph.

* remove argument.

* fix example.

* make GraphPartitionBook an abstract class.

* fix.

* fix.

* fix a bug.

* fix a bug in example

* fix a bug

* reverse heterograph sampling.

* temp fix.

* fix lint.

* Revert "temp fix."

This reverts commit c450717b9f578b8c48769c675f2a19d6c1e64381.

* compute norm.

* Revert "reverse heterograph sampling."

This reverts commit bd6deb7f52998de76508f800441ff518e2fadcb9.

* fix.

* move id_map.py

* remove check

* add more comments.

* update docstring.

Co-authored-by: Ubuntu <ubuntu@ip-172-31-2-202.us-west-1.compute.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-9-132.us-west-1.compute.internal>
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-62-171.ec2.internal>
2021-01-24 19:18:00 -08:00
Da Zheng 75ffc31f76 [Doc] Update the docstring of distributed APIs. (#2025)
* add doc.

* update DistGraph.

* add DistTensor.

* update DistEmbedding.

* add partition.py

* add sampling.

* fix.

* add graph partition book and create a base class.

* fix test.

* add rst.

* update doc rst.

* update.

* fix.

* fix docs

* update distributed tensor and embeddings.

* add checks.

* update DistGraph.

* update initialization.

* fix graph partition book.

* update graph partition book.

* update partition.

* update partition.

* fix.

* add example code.

* update DistGraph

* Update python/dgl/distributed/dist_context.py

Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* Update python/dgl/distributed/dist_context.py

Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* Update python/dgl/distributed/dist_dataloader.py

Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* Update python/dgl/distributed/dist_dataloader.py

Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* Update python/dgl/distributed/dist_dataloader.py

Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* update initialize.

* update dataloader.

* update distgraph.

* update DistGraph.

* update DistTensor.

* update.

* more updates.

* fix lint.

* add num_nodes and num_edges

Co-authored-by: Chao Ma <mctt90@gmail.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>
2020-08-18 11:53:45 -07:00
Da Zheng d1cf5c383e [Distributed] adjust various APIs. (#1993)
* rename get_data_size.

* remove g from DistTensor.

* remove g from DistEmbedding.

* clean up API of graph partition book.

* fix DistGraph

* fix lint.

* collect all part policies.

* fix.

* fix.

* support distributed sampler.

* remove partition.py
2020-08-11 20:22:00 -07:00
Da Zheng 562871e76b [Distributed] add distributed evaluation. (#1810)
* add eval.

* extend DistTensor.

* fix.

* add barrier.

* add more print.

* add more checks in kvstore.

* fix lint.

* get all neighbors for eval.

* reorganize.

* fix.

* fix.

* fix.

* fix test.

* add reuse_if_exist.

* add test for reuse_if_exist.

* fix lint.

* fix bugs.

* fix.

* print errors of tcp socket.

* support delete tensors.

* fix lint.

* fix

* fix example

Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-1.us-west-2.compute.internal>
2020-07-22 00:40:14 -07:00
Da Zheng ea420c0abd [Distributed] Reduce memory consumption in graph partitioning (#1823)
* save mem.

* save mem.

* reduce mem

* fix test

* fix lint

* fix test

* fix.

* fix.

* fix.

* fix.

* fix lint.

* fix backend operator.

* fix tensorflow operators.

* fix.

* revert change in mxnet operator.

Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-1.us-west-2.compute.internal>
2020-07-18 06:53:48 -07:00
Da Zheng 6a3685bea0 [BUGFIX] fix a bug in graph partitioning. (#1769)
* fix

* use utils.toindex in the right place.

* fix.

* update tensor for mxnet backend.

* fix

* fix
2020-07-09 01:16:42 -07:00
Da Zheng cadcc1c2a6 [Feature] add sparse embedding. (#1497)
* add sparse embedding.

* fix

* add test.

* man fixes.

* many fixes

* fix sparse emb.

* fix.

* fix lint.

* fix lint.

* fix kvstore.

* expose DistTensor.

* test sparse embeddings.

* add attach_grad to the backends.

* remove part_id

* fix.

* move backward computation.

* move more computation to backend.

* fix a bug when applying learning rate.

* fix a few things.

* fix a few things.

* add docstring

* fix.

* apply no_grad.

* fix tests.

* fix for other frameworks.

* add examples in docstring.
2020-07-03 10:42:09 -07:00
Da Zheng fa1478a4bd [BUGFIX] fix a bug in partition book. (#1698)
* fix a bug.

* add msg for assert
2020-06-24 12:38:58 -07:00
Jinjing Zhou fb248b678a [RPC] Sampling over RPC (#1616)
* fix

* test1111

* 111

* 111

* fff

* lint

* 111

* lint

* lint

* 111

* fijx

* 111

* fix

* 111

* commit

* 111

* 111

* lint

* fix typo

* fix

* lint

* fix

* 111

* support mxnet

* support mxnet

* lint

* remove print

* fix

* fix test.

* fix test.

* fix test.

* try to fix an undetermistic error.

Co-authored-by: Chao Ma <mctt90@gmail.com>
Co-authored-by: Da Zheng <zhengda1936@gmail.com>
2020-06-18 12:05:23 -07:00
Da Zheng ff5b5a4a10 [Distributed] support no shared memory. (#1663)
* support no shared memory.

* add test.

* add CAPI to check existence of shared memory.

* revert the change in src/runtime/ndarray.cc

* update docstring.

* fix compile.
2020-06-18 09:42:11 -07:00
Da Zheng 41349dcef5 [Feature] Range partition (#1522)
* add reorder immutable graph.

* add python API.

* add reorder for csr.

* remove gk version.

* fix

* add cpp test.

* bug fixes

* fix tests.

* fix bugs and add check

* fix test.

* add omp.

* add comments.

* add coo reorder.

* fix a bug.

* handle reorder for different graph structues.

* fix lint.

* fix.

* add original ids.

* reshuffle nodes before metis partition.

* inner nodes are in contiguous Id range.

* reshuffle nodes/edges when partitioning.

* load partition return graph partition book.

* use inner_node/inner_edges

* add and test range partition book.

* count inner_edge correctly.

* fix lint.

* fix lint.

* fix lint.

* fix errors.

* fix errors.

* fix for TF.

* fix.

* fix.

* change docstring.

* support logical and.

* add comments.

* avoid copy.

* fix

* update docstring.

* fix a bug.

* add range search.

* fix

* fix a bug.

* add more tests.

* load graph partition book.

* support shared memory for range partition book.

* fix a bug.

* fix.

* fix lint.

* remove check

* fix test.

* remove num_nodes and num_edges

* fix lint.

* fix graph partition book.

* address comments.

* use makedirs.

* fix compile

Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>
Co-authored-by: Chao Ma <mctt90@gmail.com>
2020-06-15 11:43:04 -07:00
Da Zheng 50962c433f Refactor distributed graph store (#1597)
* refactor graph store for new kvstore.

* fix kvstore.

* fix.

* fix lint complains.

* add docstring.

* small fix in graph partition book.

* fix.

* remove tests

* disable sampling.

* Revert "disable sampling."

This reverts commit 70451008f61ff1481d5dadbf8bd199470aee559d.

* Revert "remove tests"

This reverts commit 1394364243bdd73b669abde6193a74e2cda5521d.
2020-06-10 18:51:32 -07:00
Chao Ma 64f49703af [KVStore] Re-write kvstore using DGL RPC infrastructure (#1569)
* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update init_data

* update server_state

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* debug init_data

* update

* update

* update

* update

* update

* update

* test get_meta_data

* update

* update

* update

* update

* update

* debug push

* update

* update

* update

* update

* update

* update

* update

* update

* update

* use F.reverse_data_type_dict

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* fix lint

* update

* fix lint

* update

* fix lint

* update

* update

* update

* update

* fix test

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* set random seed

* update
2020-06-04 16:48:53 +08:00
Da Zheng 96984faca4 [Feature] Split data for distributed training (#1505)
* fix tests in graph partition book.

* implement node_split and edge_split.

* fix a bug.

* add tests.

* remove unnecessary code.

* avoid storing graph structure in partition book.

* add partition book in DistGraph.

* Revert "avoid storing graph structure in partition book."

This reverts commit 55b11fbf9293efcae5f8c97e93f1e121399dadae.

* small fixes.

* add a test for boolean mask vector.

* fix test.

* fix bugs.

* make it work for all different cases.

* fix tests.

* fix boolean mask

* fix for TF

* fix for tensorflow.

* fix test for TF

* only support boolean mask for now.

* fix tests.

* make the code more readable.

* fix test.

Co-authored-by: Chao Ma <mctt90@gmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
2020-05-14 12:46:37 -07:00
Chao Ma ef7e475051 [Distributed] Add GraphPartitionBook (#1496)
* Add graph-partition-book

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* udpate

* fix lint

* update

* update

* update

* update

* update

* update

* update

* skip test on GPU

* update

* update

* update

* update

* update

* update

* update

* fix lint

* update

* update

* update

* update

* fix unittest

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update
2020-05-08 14:52:12 +08:00