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/2799#issuecomment-539352070,https://api.github.com/repos/pydata/xarray/issues/2799,539352070,MDEyOklzc3VlQ29tbWVudDUzOTM1MjA3MA==,9155111,2019-10-08T06:08:27Z,2019-10-08T06:08:48Z,CONTRIBUTOR,I suspect [system jitter in the profiling](https://github.com/pydata/xarray/issues/2799#issuecomment-538570946) as the time for `Dataset.isel` went up. It would be useful to run [`sudo python -m pyperf system tune`](https://pyperf.readthedocs.io/en/latest/cli.html#pyperf-system) before running profiler/benchmarks.,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,416962458
https://github.com/pydata/xarray/issues/2799#issuecomment-538366978,https://api.github.com/repos/pydata/xarray/issues/2799,538366978,MDEyOklzc3VlQ29tbWVudDUzODM2Njk3OA==,9155111,2019-10-04T11:57:10Z,2019-10-04T11:57:10Z,CONTRIBUTOR,"> At first sight it looks somewhat like a hybrid between Cython (for the ahead-of-time transpiling to C++) and numba (for having python-compatible syntax).

Not really. Pythran always releases the GIL and does [a bunch of optimizations between transpilation and compilations](https://serge-sans-paille.github.io/pythran-stories/being-more-than-a-translator.html). 

A good approach would be try out different compilers and see what performance is obtained, without losing readability (https://github.com/pydata/xarray/issues/2799#issuecomment-469444519). See scikit-image/scikit-image/issues/4199 where the package `transonic` was being experimentally tested to replace Cython-only code with python code + type hints. As a bonus, you get to switch between Cython, Pythran and Numba,","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,416962458