diff --git a/README.md b/README.md index ff3b923..14d4733 100644 --- a/README.md +++ b/README.md @@ -20,7 +20,7 @@ The goals of the ML Code Completeness Checklist are to: We find that those repositories that score higher on the checklist tend to have a higher number of GitHub stars. -We've verified this by analysing NeurIPS 2019 repositories. For more details on this analyis please refer to our [blog post](https://medium.com/paperswithcode/). We also provide the [data](notebooks/code_checklist-neurips2019.csv) and [notebook](notebooks/code_checklist-analysis.Rmd) to reproduce this analysis from the post. +We've verified this by analysing NeurIPS 2019 repositories. For more details on this analyis please refer to our [blog post](https://medium.com/paperswithcode/). We also provide the [data](notebooks/code_checklist-neurips2019.csv) and [notebook](notebooks/code_checklist-analysis.pdf) to reproduce this analysis from the post. The checklist is made to be as general as possible. It consists of five items: diff --git a/notebooks/code_checklist-analysis.Rmd b/notebooks/code_checklist-analysis.Rmd index a3eac9e..adb4cf1 100644 --- a/notebooks/code_checklist-analysis.Rmd +++ b/notebooks/code_checklist-analysis.Rmd @@ -64,10 +64,10 @@ layout(matrix(c(1,2), 1, 2, byrow = TRUE), widths=c(3,2)) barplot(medians, xlab="", ylab="Median GitHub stars", ylim=c(0,200), - col=brewer.pal(6, "Blues"), cex.axis=0.7, cex.names=0.7) + col=brewer.pal(6, "Blues"), cex.axis=0.6, cex.names=0.6) mtext("GitHub repos grouped by number of ticks on ML code checklist", side=1, line=3, cex=0.8) -pie(rev(props), col=rev(brewer.pal(6, "Blues")), cex=0.7) +pie(rev(props), col=rev(brewer.pal(6, "Blues")), cex=0.6) mtext("Proportion of repositories in each group", side=1, line=3, cex=0.8) ``` diff --git a/notebooks/code_checklist-analysis.pdf b/notebooks/code_checklist-analysis.pdf index 01c95d1..11626b5 100644 Binary files a/notebooks/code_checklist-analysis.pdf and b/notebooks/code_checklist-analysis.pdf differ