4 ms·
A couple recommendations for your visualization: 1) More fine-grained bins to help show the shape of the distribution (are there performance cliffs?). Try usin
by Espressosaurus 3y ago
A couple recommendations for your visualization:
1) More fine-grained bins to help show the shape of the distribution (are there performance cliffs?). Try using vertical lines to denote % cutoffs.
2) Given the wide range between your bins, a log scale might be a good idea instead of raw frequency.
3) Try some other method of visualization. I'm not sure a histogram is useful for what you're trying to convey, at least the way it's being used here.
As it stands, the visual information is so dominated by the 99.5% case that the plots don't help illustrate your tabular data.
- wyldfire 3y agoAgreed, these graphs were challenging to read. Instead of a histogram, a violin plot (especially "split") might be an excellent way to compare the distributions. And that solves the bucket size problem too IMO. https://seaborn.pydata.org/generated/seaborn.violinplot.html https://seaborn.pydata.org/generated/seaborn.violinplot.html