Bug with y_to_force

[sginot]

Hi Peter

When using function y_to_force() with a classifier tibble that contains the "required" columns, including a species column, the output tibble does not include all columns in the classifier, but only the "measurement", "specimen", "t" and "force" columns. Notably, the species column seems to be dropped.

Then, if trying to use function summarize_measurements on the output tibble, the follwing error is issued:

Error in summarize_measurements(df.all.200.tax, var1, var2) : 
  column names of 'df' must contain 'species', 'specimen', 'measurement', 't', 'force'

Since only the species column is missing, I guess that's where the error stems from.

Cheers

Sam

[Peter-T-Ruehr]

Hi Sam,

thanks for bringing this up! I updated the y_to_force() function so that the classifier only needs the columns "measurement", "amp", and "lever.ratio". All other columns (e.g., "specimen", "species") are not necessary and may differ from user to user and from project to project. Since the example of the vignette uses these columns later, I added the following command to the vignette after running y_to_force() to get the columns according to the classifier into the df.all.200.tax data frame:

# add species and specimen info to df.all.200.tax df.all.200.tax <- df.all.200.tax %>% left_join(classifier %>% select(species, specimen, measurement))

I hope this helps!
Cheers,
Peter

[sginot]

Hi again!

Thanks for this. Follow-up question: isn't there a similar problem with function find_strongest_peaks() ? When excluding the species column from my tibble, an error is also issued:
Error in distinct(): ! Must use existing variables. x species not found in .data.

However, one might want to find strongest peaks per measurement or per individual. My current workaround is just to replace the content of the "species" column by whichever grouping factor I want to use, but I guess it might be confusing for new users ;)