Why Seaborn's histogram plot takes a long time to plot when the data is far from the median than when it is closer?

Seaborn's histogram plot takes longer to plot when the data is far from the median because the default algorithm used by Seaborn to calculate the bin size is based on the Freedman-Diaconis rule, which considers the interquartile range (IQR) of the data to calculate the bin size.

When the data is far from the median, the IQR is larger, which means that the bin size is larger, and therefore more data points are likely to fall into each bin. This can make the computation of the histogram take longer, as Seaborn needs to count the number of data points in each bin.

On the other hand, when the data is closer to the median, the IQR is smaller, and the bin size is also smaller. This means that fewer data points are likely to fall into each bin, making the computation of the histogram faster.

It's also worth noting that Seaborn's histogram plot has a parameter called bins that allows you to manually set the number of bins used to create the histogram. If you set the number of bins to a smaller value, the computation time for the histogram will decrease, regardless of where the data is relative to the median. However, using a smaller number of bins can result in a less detailed and less accurate histogram.