项目文件夹

文件

T

Da Zheng 729ff2ef38 Fix the performance issue of graph partitioning in new DGLGraph (#1934 )

* fix perf.

* fix.

* accelerate metis.

* fix lint.

* use gklib.

* fix perf.

* fix.

* update metis.

* update launch script

* handle synchronized API.

* fix.

* fix example.

* fix dataloader.

* temp fix.

* temp fix omp.

* distinguish roles.

* initialize iterator of DistDataloader correctly.

* check the correctness of launch script.

* move feature copy to sampler.

* measure mem/network copy time.

* remove

* Revert "measure mem/network copy time."

This reverts commit 86cefdc14b7815fcf5aad6496af912dba48e4aa6.

* fix.

* fix

* fix.

* fix cmake.

* disable metis in windows.

* disable metis tests in windows.

* remove test for multigraph.

* fix test.

* fix.

* fix cmake.

* fix.

* revert.

Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-115.us-west-2.compute.internal>
Co-authored-by: Ubuntu <ubuntu@ip-172-31-19-1.us-west-2.compute.internal>

2020-08-09 18:33:05 -07:00

cub @ c3cceac115

[CUDA][Kernel] More CUDA kernels; Standardize the behavior for sorted COO/CSR (#1704 )

2020-06-28 18:37:28 +08:00

dlpack @ bee4d1dd8d

import ffi solution from TVM

2018-09-05 10:51:31 -04:00

dmlc-core @ 7ce90a342b

[kvstore] Performance improvement for distributed kvstore (#972 )

2019-11-04 18:27:00 +08:00

googletest @ f71fb4f9a9

[Test] Add gtest to project (#547 )