Repository navigation
Plot retrospectives - #509
Conversation
pass a list of model files and get back a single tibble object of all models concatenated together
…ppended to converted outputs
Test against two SS3 models and a combination of SS3 and Rceattle models.
Multimodel data processing is now handled natively by `convert_output`
Verify that `filter_data` works on two models and that ` plot_spawning_biomass` generates a plot with two lines.
…rs alongside multiple models
…ner1/stockplotr into feat-multi-model-processing
wrapper function for producing a plot of retrospective peels for any quantity with an existing plot_* function
…ing function calls
…o plot_retrospectives
…s Rho External calcualtions allowed only so as to not enforce methodology (definition of mohn's rho changes by science center).
|
@sbreitbart-NOAA regarding our conversation yesterday about captions/alttext for retrospectives, the current way of using I can throw a CLI message if |
I'm considering a different idea: provide some semblance of a caption/alt text (not the typical data-driven ones- a static one, without placeholders, for now). I think that would help ensure the pipeline still adds a caption and alt text object to the rda, and to the asar report, which the users can (and should) edit. That way, they'll have something to edit, vs. starting from scratch. If we do this, we'd add a caption and alt text to the template csv. Then, we'd alert the user that they should update the caption and alt text for accuracy and descriptiveness until we develop functionality to create data-driven versions. Thoughts, @Schiano-NOAA ? |
Okay. This would alleviate the immediate need to handle multi-model key quantities at least. For something like retrospectives, where one function can make a bunch of different plots depending on the quantity, were you thinking of just having one "retrospectives" static captions/alt text, or a unique one for each possible quantity? |
Ultimately, the captions and alt text will need to be differentiated in some way, between plots, as they're describing different plots. However, for now, I think it's alright to export identical captions/alt text if we're signalling that the underlying architecture is still "under construction". |
|
Tuning in: I think that could be fine for an intermediate but I do think we will need to have some type of data-driven caption and alt text. I think it's actually a really important and defining part for stockplotr. The issue with pushing this to later is that we would need to keep a detailed list of what doesn't have it yet and what needs to be changed along with having to go back and do it for whatever we want. I don't think we should halt progress for this, but it will need to be done sooner rather than later so any new plots will be able to make the same approach. Admittedly, I think this is easier of a thing to see in action than describing so lmk if I am understanding incorrectly. |
|
@Schiano-NOAA I mostly agree. I think that, since this is the first plot that is specifically multi-model, its a good test case for the new caption alt-text workflow. Merging it now would make it available for thorough testing within the new workflow whenever that is implemented. I'm thinking about what other plots would explicitly require complex captions/alt-text like this by default, and Im struggling to come up with additional examples. Other plots that are reliant are multiple models (like likelihood profile plots), wouldn't require the same level of data to write a descriptive captions/alt-text. |
|
Ah okay I see. I agree this should be merged once ready and the workflow be worked out in another PR |
|
Also I am currently reviewing the PR and should have an approval or edits soon |
|
@Schiano-NOAA I think this is good to go now |




What is the feature?
plot_*function already exists (currently this isplot_biomass,plot_spawning_biomass,plot_fishing_mortality,plot_recruitment,plot_spr).How have you implemented the solution?
plot_retrospectivesfunction acts mostly as a data formatting wrapper that allows users to specify how many years to peel off the terminal data year for each model. Points on the terminal data year can also be added to the plot, akin to the plots generated in r4ss.Does the PR impact any other area of the project, maybe another repo?
PR contribution checklist
Thank you for contributing to stockplotr! Please check the boxes below to confirm this pull request (PR) is ready for review. Edit this PR description to add the release indicator and any explanations.