Datasets

[holyglenn]

Hi, I am very interested in the benchmarks, and find it very useful in knowing dask and RAPIDS. Thanks for the work.

Could you provide more guides in how to set it up?

• the referred datasets, including nyc taxi, mortgage etc.? So that the tests can be reproducible?
• the dependencies

Best,
Weiren

pybench/pybench/benchmarks/benchmark_dataframe.py

Line 31 in 67b4a8d

"/datasets/nyc_taxi/2015/yellow_tripdata_2015-01.csv",

pybench/pybench/benchmarks/benchmark_ml.py

Line 93 in 67b4a8d

@pytest.mark.parametrize("data", ["data/mortgage.npy.gz"])