search

open-numbers / ddf--gapminder--co2_emission

0 stars 5 forks source link

Autogenerated #3

Closed semio closed 4 years ago

semio commented 4 years ago

This pull request contains bug fix and update for 2 indicators:

The indicator have an issue caused by chef's window procedure (https://getsatisfaction.com/gapminder/topics/problems-with-japan-cumulative-emissions). Now the bug was fixed.

The indicator is based on CDIAC emission dataset and GM population dataset (open-numbers/ddf--gapminder--population_historic), and the data on master was calculated with earlier version of GM population data and autogenerated branch has latest version

Here is a diff report showing datapoints that changed more than 20% and additions/removals (There are no additions actually). Left side is autogenerated right side is master.

semio commented 4 years ago

@jheeffer The issue in cumulative_co2_emissions_tonnes was fixed, but running recipe also updated the emission per person indicator. The reason datapoints removed is that some geos are not available in the population dataset any more. I haven't checked in details for other diffs yet. Could you suggest how should we merge them in master?

jheeffer commented 4 years ago

What was the old population dataset in which the geo's were still present?

I see from https://github.com/open-numbers/ddf--gapminder--population_historic/commit/6934eb06747c03c0a153be7e23c199b673fc67f4 that it recently changed from v5 to v6, but I can't find many of those missing geo's in v5 either?

Edit: Ah, I guess this is where it happened https://github.com/open-numbers/ddf--gapminder--population_historic/commit/a3bdac7890dd275b21af9bfe98c962d0fda1b667

It moved from one version 5 to another, where the latter had all territories removed.

Seems Ola did it on purpose so let's just roll with it and accept deletions.