python3Packages.tensorflow-datasets: fix mlcroissant 1.1.0 compatibility
This commit is contained in:
@@ -93,6 +93,16 @@ buildPythonPackage (finalAttrs: {
|
||||
"elif field.label == field.LABEL_REPEATED:" \
|
||||
"elif hasattr(field_value, 'extend'):"
|
||||
''
|
||||
# mlcroissant 1.1.0 requires leaf fields to define `source` or `value`
|
||||
+ ''
|
||||
substituteInPlace tensorflow_datasets/core/utils/croissant_utils_test.py \
|
||||
--replace-fail \
|
||||
"references=mlc.Source(field='splits/name')," \
|
||||
"references=mlc.Source(field='splits/name'), source=mlc.Source(field='splits/name')," \
|
||||
--replace-fail \
|
||||
"references=mlc.Source(field='labels/label')," \
|
||||
"references=mlc.Source(field='labels/label'), source=mlc.Source(field='labels/label'),"
|
||||
''
|
||||
# TypeError: only 0-dimensional arrays can be converted to Python scalars
|
||||
+ ''
|
||||
substituteInPlace tensorflow_datasets/datasets/smallnorb/smallnorb_dataset_builder.py \
|
||||
|
||||
Reference in New Issue
Block a user