Just got crash while running demo while working in jupyter notebook

[XinyueZ]

import torch
from timesformer.models.vit import TimeSformer

model = TimeSformer(img_size=224, num_classes=400, num_frames=5, attention_type='divided_space_time',  pretrained_model=checkpoint_file)

dummy_video = torch.randn(2, 3, 8, 224, 224) # (batch x channels x frames x height x width)

pred = model(input_video_path,) # (2, 400)

Just crash in notebook.

While tried in "py" file, I just get "Floating point exception"

I have a video 224x224

According to may break-point debug, I stop at https://github.com/facebookresearch/TimeSformer/blob/main/timesformer/models/vit_utils.py#L47