issue_comments: 345434065

This data as json

html_url issue_url id node_id user created_at updated_at author_association body reactions performed_via_github_app issue
https://github.com/pydata/xarray/issues/1725#issuecomment-345434065 https://api.github.com/repos/pydata/xarray/issues/1725 345434065 MDEyOklzc3VlQ29tbWVudDM0NTQzNDA2NQ== 1197350 2017-11-18T10:47:48Z 2017-11-18T10:47:48Z MEMBER

Since #1532, the repr for dask-backed variables does not show any values (to avoid triggering computations). But numpy-backed lazily-masked-and-scaled data is treated differently: it is shown.

This highlights an important difference between how LazilyIndexedArray and dask array work: with dask, either you compute a whole chunk or you compute nothing. With LazilyIndexedArray, you can slice the array however you want and only apply mask_and_scale to the specific items you have selected. This small difference has big performance implications, especially for the "medium sized" datasets @benbovy refers to. If we changed so that decode_cf used dask, you would have to compute the whole chunk in order to see the repr.

So on second thought, maybe the system we have now is better than using dask for "everything lazy."

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   274797981