read dataset fill value question

[Matthew Gidden]

Hi folks,

Had a bit of a counterintuitive result today. Before making an issue, I thought I'd ask the list.

I have a nc file with _FillValue=0. When I read that in with xarray, the fillvalue used appears to by np.nan. Is that expected behavior?

Gist available here: https://gist.github.com/gidden/fc8eb41c1117f144d1e656f4800c2b82

Cheers,

Matt

[Stephan Hoyer]

Yes, by default xarray converts data points equal to _FillValue to NaN, which we treat as a missing value for all operations. You can disable this with mask_and_scale=False in open_dataset.

--
You received this message because you are subscribed to the Google Groups "xarray" group.
To unsubscribe from this group and stop receiving emails from it, send an email to xarray+unsubscribe@googlegroups.com.
To post to this group, send email to xar...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/xarray/CAHRdPFwPnwDe4dWYNHM1PXqgCNGTzgiuzQimsZ3nQLXDacduvQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[Matthew Gidden]

Ok, great. I saw this option but was a little confused with why I would also want to scale. Would a PR that added simply a `mask` or `fillna` argument to open_dataset be of interest? This is what I would have wanted for my use case, but perhaps I'm misunderstanding the design purpose.

To view this discussion on the web visit https://groups.google.com/d/msgid/xarray/CAEQ_TvcNcWStACZbOcEun%3DW72x%2BDGyshZvzJ728QeKjb7qg83Q%40mail.gmail.com.

[Stephan Hoyer]

Yes, it probably makes sense to split mask_and_scale into separate options. Though we do want to figure out a way to reduce the proliferation of these options (https://github.com/pydata/xarray/issues/939).

To view this discussion on the web visit https://groups.google.com/d/msgid/xarray/CAHRdPFwCDBOY76ghHqu1mEns-YCUV%3D2R6VQGyWSEuCJA7A%2BHbg%40mail.gmail.com.