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/791#issuecomment-195616194,https://api.github.com/repos/pydata/xarray/issues/791,195616194,MDEyOklzc3VlQ29tbWVudDE5NTYxNjE5NA==,1217238,2016-03-12T00:33:11Z,2016-03-12T00:33:11Z,MEMBER,"Why not make a PR to add `nancumsum` and `nancumprod` to NumPy as well? Then it's pretty clear that a back-port in `npcompat.py` is appropriate. That's actually how `nanprod` ended up in NumPy :).
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,140214928
https://github.com/pydata/xarray/issues/791#issuecomment-195446904,https://api.github.com/repos/pydata/xarray/issues/791,195446904,MDEyOklzc3VlQ29tbWVudDE5NTQ0NjkwNA==,1217238,2016-03-11T16:47:24Z,2016-03-11T16:47:24Z,MEMBER,"`cumsum`/`cumprod` will need a slightly different (simpler) interface than the other reduce methods, because unlike other aggregation functions they don't remove a dimension (the result has the same size as the input). Also, as you point out, NumPy doesn't have a nan-skipping version of these functions.

There was no particular reason why I didn't add these before -- we just never had a compelling enough need to get around to it. I don't think it would be particularly difficult.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,140214928