SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Remove trust_remote_code when loading Libri Dummy (#31748)

Browse Source 
* [whisper integration] use parquet dataset for testing

* propagate to others

* more propagation

* last one
This commit is contained in:
Sanchit Gandhi
2024-07-23 14:54:38 +08:00
committed by GitHub
parent 3aefb4ec7f
commit f83c6f1d02
56 changed files with 110 additions and 254 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/models/seamless_m4t/test_feature_extraction_seamless_m4t.py
View File

@@ -258,9 +258,7 @@ class SeamlessM4TFeatureExtractionTest(SequenceFeatureExtractionTestMixin, unitt
self.assertTrue(pt_processed.input_features.dtype == torch.float32)
def _load_datasample(self, id):
ds = load_dataset(
"hf-internal-testing/librispeech_asr_dummy", "clean", split="validation", trust_remote_code=True
)
ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
# automatic decoding with librispeech
speech_sample = ds.sort("id")[id]["audio"]["array"]