File tree Expand file tree Collapse file tree 2 files changed +11
-6
lines changed Expand file tree Collapse file tree 2 files changed +11
-6
lines changed Original file line number Diff line number Diff line change @@ -183,6 +183,7 @@ def get_dataset_repr() -> str:
183
183
f' explicitly set. Wrong arguments for { get_dataset_repr ()} '
184
184
)
185
185
builder_kwargs ['data_dir' ] = gcs_utils .gcs_path ('datasets' )
186
+
186
187
if name .namespace :
187
188
if name .namespace == 'huggingface' :
188
189
return huggingface_dataset_builder .builder (
@@ -827,12 +828,8 @@ def data_source(
827
828
""" # fmt:skip
828
829
builder_kwargs = _set_file_format_for_data_source (data_dir , builder_kwargs )
829
830
_validate_file_format_for_data_source (builder_kwargs )
830
- dbuilder = _fetch_builder (
831
- name ,
832
- data_dir ,
833
- builder_kwargs ,
834
- try_gcs ,
835
- )
831
+
832
+ dbuilder = _fetch_builder (name , data_dir , builder_kwargs , try_gcs )
836
833
_download_and_prepare_builder (dbuilder , download , download_and_prepare_kwargs )
837
834
return dbuilder .as_data_source (
838
835
split = split , decoders = decoders , deserialize_method = deserialize_method
Original file line number Diff line number Diff line change @@ -180,6 +180,14 @@ def get_dataset_dir(
180
180
version : version_lib .Version | str | None = None ,
181
181
) -> epath .Path :
182
182
"""Returns the data directory for the given dataset."""
183
+ # If the data dir is a TF Hub path, use the TF Hub dataset directory.
184
+ if tfhub_path .is_tfhub_filepath (str (data_dir )):
185
+ return tfhub_file_utils .get_tfhub_dataset_dir (
186
+ dataset_name = builder_name ,
187
+ config_name = config_name ,
188
+ version = version ,
189
+ tfhub_env = tfhub_file_utils .TfHubEnv .PROD ,
190
+ )
183
191
dataset_dir = epath .Path (data_dir ) / builder_name
184
192
if config_name :
185
193
dataset_dir /= config_name
You can’t perform that action at this time.
0 commit comments