Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Error - data type 'int64 for float64'

See original GitHub issue

Hi, After multiple testing (v0.7.6), it seems that the cooler load and zoomify do not expect to deal with float values. ValueError: cannot safely convert passed user dtype of int64 for float64 dtyped data in column 2

It means that only raw Hi-C counts can be loaded, but not normalized or transformed counts (pearson correlation maps, observed/expected, etc.) It would be great if future versions of cooler could fixed that to allow users to play with any type of data transformation and see them into HiGlass. Best N

Issue Analytics

State:
Created 5 years ago
Comments:8 (5 by maintainers)

Top GitHub Comments

1reaction

danydoerrcommented, Feb 15, 2019

Thanks, works!

0reactions

nvictuscommented, Feb 15, 2019

Thanks @danydoerr. My mistake, I thought you were inputting a BG2-type file, where count is in column 7, but you were using a COO file. If you do --field count=3:dtype=float, it will work. In fact, because count is standard you can omit the column number so --field count:dtype=float will work too.