Skip to main content
Data Science

Return to Answer

added 107 characters in body
Source Link
WBM
WBM
  • 707
  • 5
  • 16

You might want to look at these kinds of error band line plots in seaborn:

enter image description here

https://seaborn.pydata.org/examples/errorband_lineplots.html

You could compute this for each colour/ day of the week. The central line would be a mean value of each colour/day, the shadow being the distribution.

Then it's probably best to look at days individually to compare the distributions of functions. A box-whisker plot might be good to compare distributions, although this is typically only for 1-D data.

EDIT

You can limit the y axis with plt.ylim(0.25,1) but the previous might look more appealing

You might want to look at these kinds of error band line plots in seaborn:

enter image description here

https://seaborn.pydata.org/examples/errorband_lineplots.html

You could compute this for each colour/ day of the week. The central line would be a mean value of each colour/day, the shadow being the distribution.

Then it's probably best to look at days individually to compare the distributions of functions. A box-whisker plot might be good to compare distributions, although this is typically only for 1-D data.

You might want to look at these kinds of error band line plots in seaborn:

enter image description here

https://seaborn.pydata.org/examples/errorband_lineplots.html

You could compute this for each colour/ day of the week. The central line would be a mean value of each colour/day, the shadow being the distribution.

Then it's probably best to look at days individually to compare the distributions of functions. A box-whisker plot might be good to compare distributions, although this is typically only for 1-D data.

EDIT

You can limit the y axis with plt.ylim(0.25,1) but the previous might look more appealing

Source Link
WBM
WBM
  • 707
  • 5
  • 16

You might want to look at these kinds of error band line plots in seaborn:

enter image description here

https://seaborn.pydata.org/examples/errorband_lineplots.html

You could compute this for each colour/ day of the week. The central line would be a mean value of each colour/day, the shadow being the distribution.

Then it's probably best to look at days individually to compare the distributions of functions. A box-whisker plot might be good to compare distributions, although this is typically only for 1-D data.