calico / basenji

Sequential regulatory activity predictions with deep convolutional neural networks.
Apache License 2.0
410 stars 126 forks source link

Target Description #104

Closed khoangothe closed 2 years ago

khoangothe commented 2 years ago

Hi! Thank you for the awesome work. I am trying to apply your preprocessed data to solve another problem.

After I tried converting the tfrecord files of human data to numpy array using your code, each sample in the target has a dimension of (896, 5313). Is the order of those 5313 features consistent to the order of the identifiers listed in https://github.com/calico/basenji/blob/master/manuscripts/cross2020/targets_human.txt ?

Thank you so much.

davek44 commented 2 years ago

Yes, the order should be consistent.

khoangothe commented 2 years ago

That's great! Thank you again for the great project.