Percentile

Percentals indicate the value below which a certain percentage of the data is; for example, the 70th percentile means 70 percent of the data is worth below. NumPy provides np.percentile() that can also receive a list of percentiles so that quarts 25, 50, and 75 are counted at once in a single call. The 50th percentile is identical to the median. This concept is widely used to understand data distribution and to define the limit of compression.