GitLab

[Performance][Feature] Add `src_nodes` paramter to `to_block()` to avoid cost running unique() when available. (#2973)

* Add lhs_nodes are paremeter to to_block

* Update unit test

* Switch to simplified node conversion

* Switch lhs_nodes to be in/out parameter

* Update docs
Co-authored-by: Da Zheng <zhengda1936@gmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

* [Performance] improve coo2csr space complexity when row is not sorted

* [Perf] replace std::vector<> by NDArray

* keep both impl of unsorted coo to csr and choose according to graph density dynamically

* refine criteria to choose btw Unsorted algos
Co-authored-by: Ubuntu <ubuntu@ip-172-31-34-27.us-west-2.compute.internal>

* fix start_idx

* fix the bug when cuda > 0
Co-authored-by: Tong He <hetong007@gmail.com>

* rgcn with new heterograph API

* apply_edge() forward for multi relation

* undoing changes from rgcn-hetero

* backward apply_edge(copy_u) added

* unittest for apply_edge(copy_e)

* Compatible with new PRs

* resolving conflict with master

* Bringing back change after resolving conflict

* minor

* minor
Co-authored-by: Israt Nisa <nisisrat@amazon.com>

* Added binary builtinMsgFunc forward() for heterograph

* Added backward for u_op_v

* Supports all binary builtin forward

* Supports binary message funcs with reduce func sum

* lint check

* removed import torch from unittest

* enabled GPU test

* lint check

* Fixed docstrings

* rename func get_hs_id

* edited comment
Co-authored-by: Israt Nisa <nisisrat@amazon.com>

* remove

* remove

* fix

* remove

* remove

Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

[Feature] Add a HINT for the per edge type sampler of heterogeneous DistGraph that highlighting the etypes are sorted already. (#3260)

* pass cpp test

* distgraph use sorted edge flag.

* lint

* triger

* update test
Co-authored-by: Ubuntu <ubuntu@ip-172-31-2-66.ec2.internal>

* Make column double indices lazy

* Copy indices to proper contexts

* Fix initialization

* Add unit test

* Fix unit test for tensorflow

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

* handle empty frontiers.

* fix lint.

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

* 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

* Update filter code

* Add unit tests

* Fixes

* Switch to indices

* Rename functions

* Fix linting

* Fix whitespace

* Add doc

* Fix heterograph

* Change workspace allocation

* Fix linting

* Fix docs in filter.py

* Add todo
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

[Tools] In `tools/launch.py`, correctly pass all DGL client/server env vars if udf is a multi-command (#3245)

* Correctly pass all DGL client/server env vars if udf is a multi-command

* Refactor to use wrap_cmd_with_local_envvars() helper fn

* Fix dist negative data loader bug

* upd

* Fix
Co-authored-by: Da Zheng <zhengda1936@gmail.com>

* Refactors torch dist launcher udf-wrap code to handle more python versions

* minor changes

* Add pytorch-direct version

* remove

* Add multi-gpu unified tensor test for pytorch

* relocate verification step to each process

* reduce number of workers

* add parameter
Co-authored-by: shhssdm <shhssdm@gmail.com>
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* make heterogeneous find_edges

* add distributed EdgeDataLoader.

* fix.

* fix a bug.

* fix bugs.

* add tests on distributed heterogeneous graph sampling.

* fix.
Co-authored-by: Zheng <dzzhen@3c22fba32af5.ant.amazon.com>

* fix.

* fix.

* fix.

* fix.

* Fix test

* Deprecate old DistEmbedding impl, use synchronized embedding impl

* Basic imple of heterogeneous on homogenenous sampling

* make pass

* Pass C++ test

* Add python test code

* lint

* lint

* Add MultiLayerEtypeNeighborSampler

* Add unitest for single machine dataloader

* Add dist dataloader test for edge type sampler

* Fix lint

* fix

* support for per etype sample

* Fix some bug and enable distributed training with per edge sample

* fix

* Now distributed training works

* turn off some mxnet

* turn off mxnet for some dist test

* fix

* upd

* upd according to the comments

* Fix

* Fix test and now distributed works.

* upd

* upd

* Fix

* Fix bug

* remove dead code.

* upd

* Fix

* upd

* Fix
Co-authored-by: Ubuntu <ubuntu@ip-172-31-71-112.ec2.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-2-66.ec2.internal>
Co-authored-by: Da Zheng <zhengda1936@gmail.com>

* 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>

[Feature][Performance][GPU] Introducing UnifiedTensor for efficient zero-copy host memory access from GPU (#3086)

* Add pytorch-direct version

* Initial commit of unified tensor

* Merge branch 'master' of https://github.com/davidmin7/dgl



* Remove unnecessary things

* Fix error message

* Fix/Add descriptions

* whitespace fix

* add unpin

* disable IndexSelectCPUFromGPU with no CUDA

* add a newline for unified_tensor.py

* Apply changes based on feedback

* add 'os' module

* skip unified tensor unit test for cpu only

* Update tests/pytorch/test_unified_tensor.py
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>

* reflect feedback
Co-authored-by: shhssdm <shhssdm@gmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>

* fix.

* fix.

* fix.

* fix.

* Fix test

* Deprecate old DistEmbedding impl, use synchronized embedding impl

* update doc
Co-authored-by: Ubuntu <ubuntu@ip-172-31-71-112.ec2.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-2-66.ec2.internal>
Co-authored-by: Da Zheng <zhengda1936@gmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* Update

* Update

* Update

* Update

* Update

* Update

* Update

* Fix lint

* Fix

* Update

* Fix test cases

* Fix

* add docstrings

* Update
Co-authored-by: Ubuntu <ubuntu@ip-172-31-28-17.us-west-2.compute.internal>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

* add left normalizer for gcn

* fix

* fixes and some bug stuff

* [Feature] apply dgl.reorder() onto several node classification datasets in DGL

* rebase on latest dgl.reorder_graph()

* [Feature] enable edge reorder in dgl.reorder_graph()

* refine doc string

* refine doc string for dgl.reorder_graph

* refine doc string further

Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

[Feature] Add Heterograph support on Python for builtin unary msg functions (copy_u, copy_e) (#2989)

* heterograph for binary func

* Added SDDMM support

* Added unittest

* added binary test cases

* unary mfuncs works

* Fixed lint err

* lint check and others

* link check

* fixed import *_hetero issue

* lint check

* replace torch with dgl backend

* lint cehck

* removed torch from test

* skip mxnet unittest

* skip gpu test

* Remove unused/duplicated code

* minor

* changed data structure of ndata and edata

* link check

* reorganized

* minor lint

* minor lint

* raise error for udf func

* lint check

* fix for CUDA 10.1

* add a note for future cross-type max/min reducing

* Add support CUDA < 11

* lint check

* tidied C code

* remove dummy GSDDMM_hetero backward implementation
Co-authored-by: Israt Nisa <nisisrat@amazon.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
Co-authored-by: Quan Gan <coin2028@hotmail.com>

* add docstring for missing API; fix some docstring

* rename apis; address comments

* fix.

* fix.

* fix.

* fix.

* Fix test
Co-authored-by: Ubuntu <ubuntu@ip-172-31-71-112.ec2.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-2-66.ec2.internal>

* fix bugs in partitioning on heterogeneous graphs.

* fix.

* fix.

* fix example.

* fix.

* fix test.

* fix.

* fix.

* fix.

* fix tests.
Co-authored-by: Ubuntu <ubuntu@ip-172-31-71-112.ec2.internal>
Co-authored-by: Zheng <dzzhen@3c22fba32af5.ant.amazon.com>

* try enable kvstore test

* fix

* fix

* seperate out kvstore test

* add comment

* Add dgl.utils.is_sorted_srcdst

* Fix linting issues

* delete blank line

* Specify datatype to index tensor in test

* Force integer conversion
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

* [Feature] Add dgl.reorder() to re-order graph according to specified strategy

* fix unit test failure for metis reorder

* fix unit test failure on mxnet_cpu

* refine unit test for dgl.reorder

* fix unit test failure on mxnet

* fix array_equal error for mxnet unit test

* fix unit test failure for mxnet

* convert metis output to numpy array explicitly
Co-authored-by: Tong He <hetong007@gmail.com>

* fix

* remove nvidiasmi

* fix

* fix docs

* fix

* fix

* 1

* fix

* remove

* skip deprecated kernel

* fix

* Revert "skip deprecated kernel"

This reverts commit c5ceb7f60dbbaf065b81cc3680757fd611d90ad3.

* fix

* csr and csc creation

* fix

* fix

* fixes to adj transpose

* fine

* raise error if indptr did not match number of nodes

* fix

* huh?

* oh
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

[Bug fix] Use shared memory for grad sync when NCCL is not avaliable as PyTorch distributed backend. (#3034)

* Use shared memory for grad sync when NCCL is not avaliable as PyTorch distributed backend.

Fix small bugs and update unitests

* Fix bug

* update test

* update test

* Fix unitest

* Fix unitest

* Fix test

* Fix

* simple update
Co-authored-by: Ubuntu <ubuntu@ip-172-31-24-212.ec2.internal>