Hi @nperez I've tried this version with R4.0 (my laptop) and R3.2 (bscearth366), and it works well in both versions. I have added reshape2::melt in some lines 173, 242 and 245, to avoid a warning message with data.table melt function. Not sure if this is the best way, but now it does not complain anymore.

Do you think we can merge this branch into master? (Maybe we could take the occasion to correct some of the improvements in issue #63 (closed), before merging, although I would not want to have compatibility break with name of parameters.)

Hi @llledo,

Yes, we can include it in the master. Could you please open a merge request?

To avoid the compatibility break, you can add the parameter 'memb_dim' set by default to 'members' (the current name used for the dimension) in the last position of the parameter allowed by PlotForecastPDF. In this way, I cannot foresee any problem with scripts using the previous version.

In any case, let me know if you prefer to merge this version or to wait for the improvements on the merge request message.

Cheers,

Núria