Open ir0nt0ad opened 4 years ago
https://github.com/USTC-Video-Understanding/I3D_Finetune/blob/c32d90f38f3c031af0850cff757bd155734954ac/finetune.py#L63
It appears that we always call process_video with is_training=True, and therefore each sample of test_dataset has _CLIP_SIZE, not _EACH_VIDEO_TEST_SIZE frames as we would like to have it.
process_video
is_training=True
test_dataset
_CLIP_SIZE
_EACH_VIDEO_TEST_SIZE
@Rhythmblue @AlexHu123
https://github.com/USTC-Video-Understanding/I3D_Finetune/blob/c32d90f38f3c031af0850cff757bd155734954ac/finetune.py#L63
It appears that we always call
process_video
withis_training=True
, and therefore each sample oftest_dataset
has_CLIP_SIZE
, not_EACH_VIDEO_TEST_SIZE
frames as we would like to have it.@Rhythmblue @AlexHu123