Expose 'raw' as an argument?

[Yaoyx]

basenji/basenji/dataset.py

Line 215 in 615b9ee

dataset = dataset.map(self.generate_parser(raw=True))

It currently returns index format for dna sequence as default. Should this be exposed in case hdf5 file has index format, which seems to be the default (

basenji/bin/basenji_data_write.py

Line 157 in 615b9ee

seq_1hot = dna_1hot_index(seq_dna) # more efficient, but fighting inertia

)

This would, for example, make the explore_model.ipynb data loading consistent with the input shape and format required for a model prediction.

[davek44]

I won't have time to look into this for awhile. It sounds like you understand what ought to happen, so feel free to submit a pull request.

[Yaoyx]

I won't have time to look into this for awhile. It sounds like you understand what ought to happen, so feel free to submit a pull request.

Sounds good. I just submitted a pull request.