d3.hist2d plugin implements a non-blocking rectangular binning. It's useful for displaying a scatterplot with millions of points, aggregating the data into a more coarse representation suitable for display.
It can be used as a module of a larger application, because it doesn't block the UI, it computes the values between each animation frame.
Inspired by the work of Mike Bostock.
http://bl.ocks.org/herkulano/4f43dbf3473dc5503052
# hist2d([data], callback)
Expects [data] to be an array of arrays: [[1,2],[3,4],...]
The callback function is called when the binning is complete. It sends the data to its parameter.
d3.hist2d()
.bins(40)
.domain([[0, 100], [0, 100]])
(data, draw);
function draw(hist) {
// hist is the binned data
}The returned data is an array of bins. Each bin contains its position in columns and rows as x and y:
- x - the column's position
- y - the row's position
Bins that are empty are omitted.
# hist2d.bins(bins)
Sets the number of columns and rows for the bins of the rectangular histogram.
The total number of bins is columns * rows === bins * bins
# hist2d.indices([x index, y index])
Defines the indices of the data in the original array. This is useful if the data has more than two values or if the values are out of order.
Example:
[
[0, 43, 12248, 30],
[1, 45, 12398, 40],
[2, 46, 12456, 50],
...
]
// we want to use [2] as x and [1] as y
d3.hist2d()
.bins(40)
.indices([2, 1])
.domain([[0, 100], [0, 100]])
(data, draw);If indices is not set it defaults to [0, 1].
# hist2d.domain([ [x domain], [y domain] ])
Sets the input domains for x and y. Expects an array with two arrays of numbers.
# hist2d.size([width, height])
[width, height] of the scatterplot. Sets the width and height of the cells size / bins.
# hist2d.size()
Returns an array with the width and height of the cells [width, height].
# hist2d.interval(interval)
The binning function is non-blocking, so the values are computed between each animation frame for 12ms by default.