Plotting Matrix Datetime index decade

#164 · open · 1 comments

View on GitHub ↗

PaulMunters

I have very long dataframes with hourly observations for extended and varying time periods. The y-axis values sometimes become unreadable due to overlapping (see example). I tried in vain many options from the pandas timeframe documentation to set freq, or periods etc. Is it possible to set a fixed number of Y-axis ticks that are evenly distributed over the axis, say 5 or 10, in combination with freq="Y". Or to use decades instead of year? ![example](https://user-images.githubusercontent.com/122528556/212050392-f9ad8394-fb30-488b-93a2-59ef3fddf165.png)

Comments

ResidentMario

The `freq` feature is generally problematic. It's not powerful enough for what it does, and should either be removed or improved. See https://github.com/ResidentMario/missingno/issues/147#issuecomment-1046136924 for a previous comment to the same effect. Until I rewrite this bit of code, you should still be able to do what you want to do by editing the plot directly using `matplotlib` commands. `missingno` outputs a `matplotlib` plot after all.