Remove label dtype/depend on BED row format

kipoi / kipoiseq

Standard set of data-loaders for training and making predictions for DNA sequence-based models.

https://kipoi.org/kipoiseq/

MIT License

77 stars 13 forks source link

Remove label dtype/depend on BED row format #48

Open stefanches7 opened 5 years ago

stefanches7 commented 5 years ago

BED columns specify needed row order, and BedDatatype class is apparently made for BED format files only. With forced label dtype, I currently have problems reading strand from column 6 - it gets converted to float with default setting (dtype = None) automatically. I propose to remove dtype specification and oblige BED columns order use.

Avsecz commented 5 years ago

What about if we have an argument: format="bed3+labels"? Alternative possible values for format would be "bed6", "bed3", ... This would make the format more clear.