I am implementing the Biwi head pose notebook in google colab using google drive for data storage. The creation of dataset using
data = (PointsItemList.from_folder(path)
.split_by_valid_func(lambda o: o.parent.name==‘13’)
.label_from_func(get_ctr)
.transform(get_transforms(), tfm_y=True, size=(120,160))
.databunch().normalize(imagenet_stats)
)
is going on endlessly. However the same notebook when run from colab without google drive is quite fast.
Any solutions??