From 450980f02f4b4f02e4ec3afd5a4fc0ac72f0e0bc Mon Sep 17 00:00:00 2001 From: Weihua Hu Date: Wed, 11 Oct 2023 20:31:51 -0700 Subject: [PATCH] update --- test/datasets/test_torch_frame_datasets.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/test/datasets/test_torch_frame_datasets.py b/test/datasets/test_torch_frame_datasets.py index 139e3dce..d592b36d 100644 --- a/test/datasets/test_torch_frame_datasets.py +++ b/test/datasets/test_torch_frame_datasets.py @@ -10,7 +10,9 @@ TaskType.REGRESSION ]) def test_dataframe_benchmark_match(task_type, scale): - # Make sure task_type, scale, idx triple map to the same underlying dataset + # Make sure task_type, scale, idx triple map to the fixed underlying + # dataset. New dataset can be appeneded, but the existing mapping needes to + # be preserved. datasets = DataFrameBenchmark.datasets_available(task_type=task_type, scale=scale) if task_type == TaskType.BINARY_CLASSIFICATION: