Hongzhi (Steve), Chen

[email protected]

Amazon Shanghai

Results 53 comments of


                                            Hongzhi (Steve), Chen

Using `tensor.untyped_storage()` in `InvertibleCheckpoint` class methods.

@dgl-bot

Using `tensor.untyped_storage()` in `InvertibleCheckpoint` class methods.

``` ================================== FAILURES =================================== _________________________ test_group_rev_res[idtype0] _________________________ idtype = torch.int32 @parametrize_idtype def test_group_rev_res(idtype): dev = F.ctx() num_nodes = 5 num_edges = 20 feats = 32 groups = 2 g =...

Using `tensor.untyped_storage()` in `InvertibleCheckpoint` class methods.

@dgl-bot

Using `tensor.untyped_storage()` in `InvertibleCheckpoint` class methods.

I am not sure, if this is not urgent, let's table it.

[Graphbolt] Support random walk for CSCSamplingGraph.

Typo, to support random walk sampling.

[example] add GGCM

@dgl-bot

GraphBolt

We will release a DGL 2.2 which supports Pytorch 2.2.2 and 2.3 in early May.

[Feature] Fixed sampler with limit on sampled nodes/edges in batch subgraph

@dgl-bot

[Feature] Fixed sampler with limit on sampled nodes/edges in batch subgraph

the code LGTM now, can you run `lintrunner -a` to fix the lint error? and can you add a unit test for your code. The test case can be put...

[Feature] Fixed sampler with limit on sampled nodes/edges in batch subgraph

@dgl-bot

‹
1
2
3
4
5
6
›