Manuscrit review LC

[LucieContamin]

The manuscrit looks great, it just wanted to propose some minor (mostly style) updates:

• I tried to follow the same "style" on all the code but might have miss some places. I mostly add some missing spaces around operators, "=", etc.; add package dependencies on the function (to harmonize the whole code, please review to be sure I assign the correct dependencies); keep the same pipe style, etc.
• I also update the plot function calls by simplifying them by moving the "data process" (filtering, etc.) of the input data outside the plotting function calls
• I also have a question about the hubverse citation, the manuscrit seems to contain 2 versions: either "(Infectious Disease Modeling Hubs 2022)" or "(https://hubdocs.readthedocs.io/en/latest/index.html)" (line 24). Is that expected?

I also have some questions about the R script evaluation_functions.R:

• it contains some functions that seem to not be used in the manuscrit (plot_evaluated_scores(), plot_wis_loc(), pairwise_comparison(). Is that expected? If so, shouldn't they be store somewhere else to avoid confusion? Also some of these functions use functions that required specific package and I cannot find which package, for example fct_relevel() or permutationTest()
• for some other function like plot_evaluated_scores_forecast_date(), the parameters of the function are organized in an order that force the either use main = ... or have 5 parameters. Is that an expected behavior?

Please feel free to let me know if any issues or questions,

[lshandross]

Thanks for all your hard work with these fixes, Lucie!

Here are my answers to your questions since they concern my sections:

• Some of the functions in the evaluation_functions.R R script were used in a previous draft of the manuscript to generate additional figures/tables, but those were removed since they didn't add much to the overall message of the paper. I can delete them.
• It is necessary for plot_evaluated_scores_forecast_date() to have the first three arguments, but not the first five. main probably could also default to NULL. Would reordering the parameters solve the issue you identified?

[LucieContamin]

Thanks for the answers! I think reordering or having a default value for main will solve the issue. Thank you!

[LucieContamin]

I just push a version with lintr updates. There is still some markers that we might want to fix:

• remove commented lines (lines 691, 728, 729 and 824)
• variable name style should match snake_case or symbols (lines 180 to 296)