dmlc--dgl
nv-dlasalle
17d604b5c7
* 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>
Unit test
Python Unittest
The code organization goes as follows:
backend: Additional unified tensor interface for supported frameworks. The functions there are only used in unit tests, not DGL itself. Note that the code there are not unit tests by themselves.compute: All framework-agnostic computation-related unit tests go there.${DGLBACKEND}(e.g.pytorchandmxnet): All framework-specific computation-related unit tests go there.graph_index: All unit tests for C++ graph structure implementation go there. The Python API being tested in this directory, if any, should be as minimal as possible (usually simple wrappers of corresponding C++ functions).lint: Pylint-related files.scripts: Automated test scripts for CI.
C++ Unittest
Compile with unittest by executing the command below
# Assume current directory is the root directory of dgl, and googletest submodule is initialized
mkdir build
cd build
cmake .. -DBUILD_CPP_TEST=1
make -j${nproc}
./runUnitTests