lowertriangle() could be faster with memory preallocation #322
Milestone
Comments
|
oh goodness. Good find @vlepori ! A PR is always welcome! |
vlepori
added a commit
to vlepori/ggally
that referenced
this issue
May 15, 2020
As per ggobi#322, rewrote the loop to first allocate the df and then fill it rather than rbind()
schloerke
pushed a commit
that referenced
this issue
May 27, 2020
As per #322, rewrote the loop to first allocate the df and then fill it rather than rbind()
|
Fixed in #328 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I was tinkering with the package functions (needed a matrix scatterplot with continuous variable color aesthetic) and saw that
lowertriangle()generates the expanded dataframes for plots by growing a df inside a nested loop.This seems inefficient and unnecessary, since we know the final dimension of the output. So I made a version of the function with memory preallocation, and it is notably faster with larger data.
The actual plotting might still take the most time, but if there is interest I'd be happy to share the code or make a PR. Thanks for the nice package!
The text was updated successfully, but these errors were encountered: