weights for 'gdp' is defined twice

[lolow]

Should we use the SSP source or the IMF source?

[JohannesEmm]

I would say WDI should be used as is what we use for our gdp calibration.
Also I cleaned up the weights in the branch https://github.com/witch-team/witchtools/tree/witch17eums

[lolow]

ok, it would be interesting to see the impact on the WITCH runs with this change in weights. I do a big refactoring in the branch convert_gdx, so the package could be outside witch.

…

On Mon, Mar 23, 2020 at 1:49 PM Johannes Emmerling ***@***.***> wrote: I would say WDI should be used as is what we use for our gdp calibration. Also I cleaned up the weights in the branch https://github.com/witch-team/witchtools/tree/witch17eums — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#6 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAG3PPWP2HJPGBFJSTUS2TRI5LF3ANCNFSM4LR2THKQ> .

[JohannesEmm]

ok great!
ONe more thing/feature request: I think would be good to have a fall back where if piggyback doesn't have the file looks at witch-data and in case finds it uses this data including printing a warning, could be good for new data asily ading it without first uploading on the release while using the standard data from github