This visualization highlights the distribution shape. All bins are the same height, regardless of their value (e.g. the bin for values between 0.1-0.2 is the same height as the one for values from 9000-10000). We’re interested in the amount of entries in a bin (shown by the color), not the value of the bin.
You can change the filename value to any of the .csv files in the gist (blocked.csv, connected.csv, duration.csv, etc)
Each column is an hour on May 1st, 2016. Each row is a bin (e.g. 1-2ms, 300-400ms). The color in each cell is scaled based on the number of entries in that bin.
Forked from mbostock‘s block: Heatmap (2D Histogram, CSV)
forked from jfsiii‘s block: Heatmap - weighted by counts