pygraphistry icon indicating copy to clipboard operation
pygraphistry copied to clipboard

[BUG] hypergraph dask_cudf exn

Open lmeyerov opened this issue 3 years ago • 0 comments

Describe the bug

dask_cudf engine failing on hypergraph unit test

To Reproduce

        edges_gdf = cudf.DataFrame({'x': ['a'], 'y': ['c']})
        n_dgdf, e_dgdf = edges_to_hypergraph(edges_gdf, {'direct': True, 'engine': 'dask_cudf'})
        #     g = graphistry.hypergraph(edges_gdf, **cleaned_opts)['graph'] ...
        n_gdf = await gpu_client.compute(n_dgdf)
        e_gdf = await gpu_client.compute(e_dgdf)

Actual behavior

____________________________________________________________________ Test_edges_to_hypergraph.test_edges_to_hypergraph_dask_dask_cudf_explicit ____________________________________________________________________

self = <test_hypergraph.Test_edges_to_hypergraph object at 0x7ff0e0160110>, gpu_client = <Client: 'tcp://172.21.0.2:8786' processes=1 threads=1, memory=7.63 GiB>

    @pytest.mark.skipif(not is_gpu(), reason='not is_gpu()')
    @pytest.mark.timeout(60)
    @pytest.mark.asyncio
    async def test_edges_to_hypergraph_dask_dask_cudf_explicit(self, gpu_client):
        edges_gdf = cudf.DataFrame({'x': ['a'], 'y': ['c']})
>       n_dgdf, e_dgdf = edges_to_hypergraph(edges_gdf, {'direct': True, 'engine': 'dask_cudf'})

test/server/client/graph/test_hypergraph.py:160: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
server/client/graph/hypergraph.py:138: in edges_to_hypergraph
    g = graphistry.hypergraph(edges_gdf, **cleaned_opts)['graph']
/conda/envs/rapids/lib/python3.7/site-packages/graphistry/pygraphistry.py:489: in hypergraph
    engine=engine, npartitions=npartitions, chunksize=chunksize)
/conda/envs/rapids/lib/python3.7/site-packages/graphistry/hyper.py:23: in hypergraph
    engine=engine, npartitions=npartitions, chunksize=chunksize)
/conda/envs/rapids/lib/python3.7/site-packages/graphistry/hyper_dask.py:739: in hypergraph
    entities = format_entities(events, entity_types, defs, direct, drop_na, engine_resolved, npartitions, chunksize, debug)  # type: ignore
/conda/envs/rapids/lib/python3.7/site-packages/graphistry/hyper_dask.py:336: in format_entities
    mt_df = mt_nodes(defs, events, entity_types, direct, engine)
/conda/envs/rapids/lib/python3.7/site-packages/graphistry/hyper_dask.py:302: in mt_nodes
    .head(0)
/conda/envs/rapids/lib/python3.7/site-packages/dask/dataframe/core.py:1049: in head
    return self._head(n=n, npartitions=npartitions, compute=compute, safe=True)
/conda/envs/rapids/lib/python3.7/site-packages/dask/dataframe/core.py:1082: in _head
    result = result.compute()
/conda/envs/rapids/lib/python3.7/site-packages/dask/base.py:284: in compute
    (result,) = compute(self, traverse=False, **kwargs)

lmeyerov avatar Jun 29 '21 23:06 lmeyerov