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/900#issuecomment-233506655,https://api.github.com/repos/pydata/xarray/issues/900,233506655,MDEyOklzc3VlQ29tbWVudDIzMzUwNjY1NQ==,1217238,2016-07-19T01:31:40Z,2016-07-19T01:31:40Z,MEMBER,"Glad to help!
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,166195300
https://github.com/pydata/xarray/issues/900#issuecomment-233504857,https://api.github.com/repos/pydata/xarray/issues/900,233504857,MDEyOklzc3VlQ29tbWVudDIzMzUwNDg1Nw==,17951292,2016-07-19T01:17:43Z,2016-07-19T01:17:43Z,NONE,"Thanks Stephan - that was very helpful! I was able to carry out the necessary computations with no problems after seeing your reply.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,166195300
https://github.com/pydata/xarray/issues/900#issuecomment-233472927,https://api.github.com/repos/pydata/xarray/issues/900,233472927,MDEyOklzc3VlQ29tbWVudDIzMzQ3MjkyNw==,1217238,2016-07-18T22:03:16Z,2016-07-18T22:03:16Z,MEMBER,"If you write a function that can take an `xarray.Dataset` as input, you can still use groupby and apply. So generally the pattern I would use is:

```
obs = obs.reindex_like(fcst)   # possibly using tolerance if the grid isn't exactly the same
combined = xarray.Dataset({'obs': obs, 'fcst': fcst})
combined.groupby('time').apply(calculate_skill)
```

It's hard to answer in more specifics without seeing your code/data, though.

If you need to do time shifts between reference time and (actual) time, it's pretty straightforward with xarray (we had some custom code to do that at my last job) but it isn't a one-liner.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,166195300