search

pfmc-assessments / VASTWestCoast

VAST for the NWFSC West Coast data
Other
5 stars 1 forks source link

Pearson residual plot by year for the Triennial #14

Closed chantelwetzel-noaa closed 5 years ago

chantelwetzel-noaa commented 5 years ago

It appears that the figures produced showing the Pearson residuals (both encounter and catch rate) are subset the incorrect years to plot. As example, when the Triennial is run as a single index there are 7 panels shown in the figures, but only 3 of them appear to be referencing years with data. Also, the "main" label for each figure is not shrunk down to fit the scale of each figure making it impossible to determine which year is being plotted. This issue is also impacting the runs creating early and late indices, but in these instances only a single year is being plotted. Here again the main label is not informative as to which year is being shown.

maps--catchrate_pearson_resid

kellijohnson-NOAA commented 5 years ago

There were two issues here, (1) I hard coded the size of the labels to wrongly be 1.8, they are now 0.8 with cbcd1b49f72b95a1ccf61aafd5470c32d28c5505 and (2) FishStatsUtils::plot_residuals did not pass Years2Include to FishStatsUtils::PlotMap, which is now fixed in james-thorson-NOAA/FishStatsUtils@f6eacfdcd127463602456669f17531a8d2dae4e4 I haven't checked this for the case where the Triennial is split, but I am assuming it will work and closing it for now.

chantelwetzel-noaa commented 5 years ago

I am still not able to get residual plots with the correct years. The residual plots that I am getting look the same to the one above. So far I have tried reinstalling the FishStatsUtils package and also passing the specific years (Year_Set) in the Sim_Settings to no avail. Please let me know if there is something that I am missing.

kellijohnson-NOAA commented 5 years ago

The fix was added to FishStatsUtils in a pull request that has not been merged by @James-Thorson yet, so temporarily you will have to install my fork.

chantelwetzel-noaa commented 5 years ago

Using your fork solves this issue. Thank you.