Issue with manually added colours

[thomadam]

I am attempting to create an animation of animal movement from a CSV file.

The file is brought into R Studio, converted to a dataframe, and duplicate data is removed.
The data is then divided into the three animal types and sex and each of the converted to a move class object using df2move().
The move object is then assigned a colour using something like AnimalTypeSex$colour = "pink".
Move objects are then converted to a moveStack using moveStack() and the times are aligned using align_move().

The issue arises when using frames_spatial() and this error pops up:
Error in $<-.data.frame(*tmp*, "tail_colour", value = c("#FFD2D2", :
replacement has 99 rows, data has 100

carFS = frames_spatial(AnimalsAligned, map_service = "osm", map_type = "terrain_bg",
trace_show = FALSE,path_fade = F, path_legend = F, path_colours = NA)%>%
add_labels(x="Longitude", y="Latitude")%>%
add_timestamps(AnimalsAligned, type="label")%>%
add_progress()

Is there a way to deal with the replacement has 99 rows, data has 100 issue?

There is no issue if the colour attribute is not used, but then all tracks are randomly coloured.

[swforrest]

I am also having this issue! Did you happen to find a solution?

[thomadam]

I ended up manually entering the colour values and was not able to use this method. It is less than ideal, but it worked in the end. It was necessary to have a correct count of the values in the moveStack(), and then manually entering the colour code in frames_spatial(path_colours = c("blue","blue","red".....))

…

On Mon, May 9, 2022 at 9:28 PM swforrest ***@***.***> wrote: I am also having this issue! Did you happen to find a solution? — Reply to this email directly, view it on GitHub <#104 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKFEXHOEKXXCWEKVN43GHQDVJHJWBANCNFSM5ICTNFGA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[swforrest]

I have found another hacky solution, but which might be a bit more straightforward to implement:
For my dataset, aligning the movement introduced NAs into the colour column, which I think would be causing the error.

I then used the 'na.locf' function from the 'zoo' package, which replaces NAs with the previous value appearing in the vector, and fills forward. This obviously assumes that the behavioural state at the NA time step is the same as the previous time step (or earlier if multiple consecutive NAs), but for something like sex then this would be fine.

I used this after the align move step:

# animating simluated data from HMM
sim1.move <- df2move(df = sim1, proj = "EPSG:4326", x = 'x', y = 'y', time = 'time.stamps', track_id = "ID")

# add the colour vector back into the data frame
sim1.move$colour <- sim1$colour 

# align_move to get move object, which keeps the colour vector but introduces NAs
move.obj <- align_move(sim1.move, res = 1, unit = "hours")

# fill NAs with the colour value that appeared previously in the vector
move.obj$colour <- na.locf(move.obj$colour)

# you can also use the argument 'fromLast = TRUE' to fill from the next value rather than the previous
# move.obj$colour <- na.locf(move.obj$colour, fromLast = TRUE)

# frames_spatial now works fine, and the colour is according to behavioural state in my example
frames <- frames_spatial(move.obj, 
                         alpha = 0, # for white background (my simulated data had no reference point)
                         path_legend = F,
                         equidistant = T,
                         trace_show = T)

Then the animate_frames also appears to work fine!