Merge pull request #60 from Genentech/tdc-dataset-fix

ChEMBL, ZINC dataset y_keys and separator
Genentech · Jan 22, 2025 · b21da66 · b21da66
2 parents 34770f9 + b6f10a6
commit b21da66
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 3 deletions.
diff --git a/src/beignet/datasets/_chembl_dataset.py b/src/beignet/datasets/_chembl_dataset.py
@@ -34,6 +34,6 @@ def __init__(
             suffix="csv",
             checksum="md5:b0cedfe468c6331487f7084236944bac",
             x_keys=["smiles"],
-            y_keys=[],
+            y_keys=None,
             transform=transform,
         )
diff --git a/src/beignet/datasets/_tdc_dataset.py b/src/beignet/datasets/_tdc_dataset.py
@@ -45,7 +45,7 @@ def __init__(
 
         match path.suffix:
             case ".csv":
-                self._data = pandas.read_csv(path)
+                self._data = pandas.read_csv(path, sep=None)
             case ".pkl":
                 self._data = pandas.read_pickle(path)
             case ".tab" | ".tsv":

diff --git a/src/beignet/datasets/_zinc_dataset.py b/src/beignet/datasets/_zinc_dataset.py
@@ -34,6 +34,6 @@ def __init__(
             suffix="tsv",
             checksum="md5:9e4754d72db297d496def3498a926979",
             x_keys=["smiles"],
-            y_keys=[],
+            y_keys=None,
             transform=transform,
         )