2 ms·
I certainly understand some arguments against rainbow, but in the same time the new default colour map in matplotlib (viridis) is (IMO) worse for many applicati
by sega_sai 5y ago
I certainly understand some arguments against rainbow, but in the same time the new default colour map in matplotlib (viridis) is (IMO) worse for many applications. It doesn't give enough contrast/doesn't define contour-lines so well as rainbow. It' true that I could probably find some other colour map, but I still often switch to jet.
- beojan 5y agoPart of the point of viridis is to not have random contours that stick out because of the colour map. If you want to highlight certain contours, plot the contour lines.
- NavinF 5y agoTurbo has all the contrast, but none of the false contours of Jet: https://ai.googleblog.com/2019/08/turbo-improved-rainbow-colormap-for.html https://ai.googleblog.com/2019/08/turbo-improved-rainbow-col...
- bluenose69 5y agoI suggest turbo to lots of people who love jet and they find it a useful replacement for those cases where the distinctness of colours is helpful, e.g. noisy fields for which contours are too much of a mess to understand. As others have said, the choice of colour scheme depends on the application. And it can even make sense to use a distinctly suboptimal scheme, if the goal is to produce graphs that can be compared easily with existing works. For example, annual reports of datasets are a lot easier to compare if the changes to the colour scheme are avoided.