GitLab

* WIP

* fix shape of stack reducer

* apply merge order by edge id for stack reducer

* WIP: working on random choices

* light slice

* basic CPU impl

* add python binding; fix CreateFromCOO and CreateFromCSR returning unitgraph

* simple test case works

* fix bug in slicing probability array

* fix bug in getting the correct relation graph

* fix bug in creating placeholder graph

* enable omp

* add cpp test

* sample topk

* add in|out_subgraph

* try fix lint; passed all unittests

* fix lint

* fix msvc compile; add sorted flag and constructors

* fix msvc

* coosort

* COOSort; CSRRowWiseSampling; CSRRowWiseTopk

* WIP: remove DType in CSR and COO; Restrict data array to be IdArray

* fix all CSR ops for missing data array

* compiled

* passed tests

* lint

* test sampling out edge

* test different per-relation fanout/k values

* fix bug in random choice

* finished cpptest

* fix compile

* Add induced edges

* add check

* fixed bug in sampling on hypersparse graph; add tests

* add ascending flag

* in|out_subgraph returns subgraph and induced eid

* address comments

* lint

* fix

* typed udf

* docstrings

* new kvstore

* 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

* test warning

* update

* update

* udpate

* update

* update

* update

* update

* small fix

* small fix

* get group count

* update

* update

* make file

* update

* use addr

* get id

* partition book

* update

* partition

* barrier

* update

* loop count

* update

* update

* update

* update

* update

* update

* update

* update

* update

* add mxnet demo

* update ip

* update

* update

* update

* random

* update

* update

* update

* update

* update

* update

* fix lint

* fix lint

* fix lint

* Several optimizations on DGL-KG:
1. Sorted positive edges for sampling which can reduce random
   memory access during positive sampling
2. Asynchronous node embedding update
3. Balanced Relation Partition that gives balanced number of
   edges in each partition. When there is no cross partition
   relation, relation embedding can be pin into GPU memory
4. tunable neg_sample_size instead of fixed neg_sample_size

* Fix test

* Fix test and eval.py

* Now TransR is OK

* Fix single GPU with mix_cpu_gpu

* Add app tests

* Fix test script

* fix mxnet

* Fix sample

* Add docstrings

* Fix

* Default value for num_workers

* Add soft relation part

* Upd

* Some fix

* upd

* Now work

* Fix TransR

* Fix eval and add some doc string

* triger

* upd

* Add some training scripts for freebase multi-gpu

* upd

* upd

* upd

* upd

* upd

* fix
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>

* fix script

* t

* fix weird bugs

* fix

* fix

* new readme

* try resize

* try center

* more imgs

* fix

* graph format

* fix lint

* lint

* fix

* unit test

* lint

* add magic num

* move serialize out of struct

* lint

* serialize

* trigger CI

* fix lint

* lint
Co-authored-by: zhoujinjing09 <zhoujinjing09@users.noreply.github.com>

* Several optimizations on DGL-KG:
1. Sorted positive edges for sampling which can reduce random
   memory access during positive sampling
2. Asynchronous node embedding update
3. Balanced Relation Partition that gives balanced number of
   edges in each partition. When there is no cross partition
   relation, relation embedding can be pin into GPU memory
4. tunable neg_sample_size instead of fixed neg_sample_size

* Fix test

* Fix test and eval.py

* Now TransR is OK

* Fix single GPU with mix_cpu_gpu

* Add app tests

* Fix test script

* fix mxnet

* Fix sample

* Add docstrings

* Fix

* Default value for num_workers

* Upd

* upd

* graph format

* fix lint

* lint

* fix

* unit test

* lint

* add magic num

* move serialize out of struct

* lint
Co-authored-by: zhoujinjing09 <zhoujinjing09@users.noreply.github.com>

* Fix

* Update

* Fix model save and load problem.

Add offline inference example

* Fix test

* Fix randomness problem between test runs

* Make rdf loader deterministic

* unit graph that prefers coo queries

* auto detect coo preference

* forgot some functions

* disable lint on detect_prefer_coo

* reorg

* change comment

* lint

* fix

* move array_utils.h to src

* compact graph impl

* fix redundant copying in idhashmap

* docstring

* moving preference detection to C

* lint

* fix unit test & address comments

* hypersparse autorestrict

* docstring & fix

* revert copyto and asnumbits

* fix stupid bug

* lint

* leave a TODO for sorted COO

* fixing same node type mapping to different id in different graphs

* addresses comments

* made induced nodes a feautre column

* lint?

* First commit

* Update

* Update splitters

* Update

* Update

* Update

* Update

* Update

* Update

* Migrate ACNN

* Fix

* Fix

* Update

* Update

* Update

* Update

* Update

* Update

* Finish classification

* Update

* Fix

* Update

* Update

* Update

* Fix

* Fix

* Fix

* Update

* Update

* Update

* trigger CI

* Fix CI

* Update

* Update

* Update

* Add default values

* Rename

* Update deprecation message

* fix.

* fix.

* trying to refactor IndexSelect

* partial implementation

* add index select and assign for floats as well

* move to random choice source

* more updates

* fixes

* fixes

* more fixes

* adding python impl

* fixes

* unit test

* lint

* lint x2

* lint x3

* update metapath2vec

* debugging performance

* still debugging for performance

* tuning

* switching to succvec

* redo

* revert non-uniform sampler to use vector

* still not fast

* why does this crash with OpenMP???

* because there was a data race!!!

* add documentations and remove assign op

* lint

* lint x2

* lol what have i done

* lint x3

* fix and disable gpu testing

* bugfix

* generic random walk

* reorg the random walk source code

* Update randomwalks.h

* Update randomwalks_cpu.cc

* rename file

* move internal function to anonymous ns

* reorg & docstrings

* constant restart probability

* docstring fix

* more commit

* random walk with restart, tested

* some fixes

* switch to using NDArray for choice

* massive fix & docstring

* lint x?

* lint x??

* fix

* export symbols

* skip gpu test

* addresses comments

* replaces another VecToIdArray

* add randomwalks.h to include

* replace void * with template

* reorg tutorials and api docs

* fix

* add env var

* Trigger CI

* simplification

* add doc

* several nn example

* appnp

* fix lint

* lint

* add dgi

* fix

* fix

* fix

* fff

* docs

* 111

* fix

* change init

* change result

* tiaocan+1

* fix

* fix lint

* fix

* fix

* Add boundary check for heterograph build with card as input

* Fix when u or v is empty

* fix test_kernel.py error print

* Revert "fix test_kernel.py error print"

This reverts commit a71c20292549c0ac62a5326c30669ca4bde8febc.

* Turn op validation check on graph and bipartite by default

* upd

* udp

* upd

* update test

* Fix the random seed and fix the _print_error

* Fix check_positive_edge_sampler sample weight.
Co-authored-by: Zihao Ye <zihaoye.cs@gmail.com>
Co-authored-by: VoVAllen <VoVAllen@users.noreply.github.com>

[Bug?] Fix non-determinism when setting the same seed across different runs with single thread (#1205)

* fix multithread random seed value

* more fix

* typo

* what did i wrote

* another fix

* another try

* fix

* rewriting with kernel functions

* revert mxnet softmax changes

* lint fix
Co-authored-by: Zihao Ye <zihaoye.cs@gmail.com>

* remote copy from parent

* upd

* make test_sampler easier to pass

* [BUG] Fix empty heterograph creation with list & tuple

* more fix

* upd

* upd

* upd

* upd

* lint

* upd

* upd

* upd

* upd

* upd
Co-authored-by: VoVAllen <VoVAllen@users.noreply.github.com>

* [Feature] Pickle support for heterograph

* fix lint
Co-authored-by: VoVAllen <VoVAllen@users.noreply.github.com>

* upd

* upd

* upd

* upd

* lint

* upd

* upd

* upd

* upd
Co-authored-by: VoVAllen <VoVAllen@users.noreply.github.com>

Fixes #1182 .

This is embarrassing.

* multi-gpu

* Pytorch can run but test has acc problem

* pytorch train/eval can run in multi-gpu

* Fix eval

* Fix

* Fix mxnet

* trigger

* triger

* Fix mxnet score_func

* Fix

* check

* FIx default arg

* Fix train_mxnet mix_cpu_gpu

* Make relation mix_cpu_gpu

* delete some dead code

* some opt for update

* Fix cpu grad update

* upd

* upd

* change download link to cdn

* Update env_var.rst

* API change of kvstore

* add demo for kvstore

* update

* remove duplicated log

* change queue size

* update

* update

* update

* update

* update

* update

* update

* update

* update

* fix lint

* change name

* update

* fix lint

* update

* update

* update

* update

* change message queue size to a python argument

* change default queue size to 2GB

* OMP_NUM_THREADS=1

* add multiple NICs support for kvstore

* test

* fix lint

* update

* update

* update

* update

* update

* update

* update

* fix lint

* fix lint

* 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

* delete msg

* clear kv msg

* update

* update

* update

* update

* update

* update

* is not None

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update