Could you provide the human 3.6 dataset which contains 2624 training samples and 1135 test samples?

[bigfeetsmalltone]

This is a very solid and interesting work. Our team would like to follow this work. However, we have difficulty in finding the human 3.6 dataset that contains 2624 training samples and 1135 test samples. We will be very appreciative of your kindness if you would like to provide this dataset.