search

nel-lab / mesmerize-core

High level pandas-based API for batch analysis of Calcium Imaging data using CaImAn
Other
60 stars 15 forks source link

allow more file formats, specific more columns in create_batch(), update tests #199

Closed kushalkolar closed 1 year ago

kushalkolar commented 1 year ago

mostly stuff for serenity compatibility

kushalkolar commented 1 year ago

I should add an extras option and put fastparquet in there if it's being maintained well.

Also, make the create and load functions just guess the reader from the file extension.

kushalkolar commented 1 year ago

so parquet doesn't work with all our datatypes, should use hdf5 which seems to work

kushalkolar commented 1 year ago

should optionally make the tests run against pickle format as well.

kushalkolar commented 1 year ago

Seems like pytables is an annoying dependency with conda, can just keep pickle in the tests and we can make a pytest option to run hdf tests that we use just on linux pip.

kushalkolar commented 1 year ago

closing, not going to develop serenity any more