I was curious if the training data was available during the submission or if we needed to upload it ourselves? I've written my code to take vcf file as input and then process it, however I'm struggling to upload the 16GB of training data for SC1 and can't seem to find any information about if the training data will be made available on the server during the upload. I was wondering if there was a solution besides doing my processing before submission. Thanks

Created by Nicholas Smith smithnickh
I'm not working on the filtered vcf files. So I was hoping that the entirety of the training data would be provided in a similar format that the validation data is. So instead of looking at /testing-data/ to find the individual vcf files (for the testing data) I was hoping a corresponding /training-data/ directory existed full of all the data given to us already. This may not exist, but I was curious if someone who knew the file structure of the submission could help. Thanks
Docker's size limit is 10GB by default, your base docker with some machine learning packages typically occupies 2-4 GB.   They provided the filtered vcf file which is of much smaller size.

Including Training Data in Submission (SC1) page is loading…