Use repo_type instead of deprecated datasets repo IDs (#19202)
* Use repo_type instead of deprecated datasets repo IDs * Add missing one in doc
This commit is contained in:
@@ -342,7 +342,9 @@ class VideoMAEModelTest(ModelTesterMixin, unittest.TestCase):
|
||||
# We will verify our results on a video of eating spaghetti
|
||||
# Frame indices used: [164 168 172 176 181 185 189 193 198 202 206 210 215 219 223 227]
|
||||
def prepare_video():
|
||||
file = hf_hub_download(repo_id="datasets/hf-internal-testing/spaghetti-video", filename="eating_spaghetti.npy")
|
||||
file = hf_hub_download(
|
||||
repo_id="hf-internal-testing/spaghetti-video", filename="eating_spaghetti.npy", repo_type="dataset"
|
||||
)
|
||||
video = np.load(file)
|
||||
return list(video)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user