Changes:

• Dropped duplicate values in datasets
• Fixed proteomics scales in datasets
• Fixed package download function (Figshare API update)

Changes to Datasets

New Datasets:

• colorectal
• liver
• novartis

Renamed Datasets:

• bladderpdo -> bladder
• pancpdo -> pancreatic
• sarcpdo -> sarcoma

Removed Datasets:

• cptac
• mpnstpdx (now part of mpnst)

Other changes:

• implemented md5 checksum verification for downloads
• added functionality to generate basic statistics for datasets

This update includes some minor fixes to the package as well as the addition of 3 new datasets:

• bladderpdo
• pancpdo
• sarcpdo

Fix to the Dataset.split_train_other() function.

This update adds the balance flag to split_train_test_validate() & split_train_other() to balance the splits if they are also stratified using a drug response metric.