Alex Barghi
Alex Barghi
You need to call `.compute()` to get results. i.e. ``` t1 = time.time() pr_df = dask_cugraph.pagerank(G).compute() t2 = time.time() ``` That error is from not properly closing the dask client....
@whuLames how large is the graph you're trying to read? It could just be larger than what two cards can support. Or you may be running into OOM in the...
@whuLames I also just noticed your question #4 - to limit the GPUs used, you can set the CUDA_VISIBLE_DEVICES environment variable. For instance, if you wanted to use GPUs 0,...
@Gibyeng you're not building `cuGraph-PyG` but still trying to run the tests for it. I think what you want to do is `cd cugraph/python/cugraph` and then `pytest`. Just running `pytest`...
Not planned; no longer using dask in this way in the GNN packages.
This bug is not critical; usage is not prevented and the performance improvement from CSC mode is marginal. This work is deferred for now.
Could be, I'll definitely test once that PR is merged.
Not sure if this is needed anymore; will revisit later
No longer needed, this doesn't make sense