Printing value in each bin in hist2d (matplotlib)

Question

I would like to create a 2d histogram, where in each bin the value represented by that bin is shown in the center of that given bin. For example a hist2d of size 5x5 would have 25 values inside the final graph. This is well doable with PyROOT, but I need to use matplotlib/pyplot here.

The following has been tried according to the first answer:

fig, ax = plt.subplots()
ax.set_aspect("equal")
hist, xbins, ybins, im = ax.hist2d(x, y, bins=(4, [1,2,3,5,10,20]))
ax.text(xbins[1]+0.5,ybins[1]+0.5, "HA", color="w", ha="center", va="center", fontweight="bold")

img = StringIO.StringIO()
plt.savefig(img, format='svg')
img.seek(0)
print("%html <div style='width:500px'>" + img.getvalue() + "</div>")

There wasn't any error message but "HA" wan't displayed in the first bin at all. I am programming this in Zeppelin, thus I need to take img from buffer ...

Answer 1

To annotate a hist2d plot, just like any other plot, you may use matplotlib's text method. The values to annotate are given by the returned histogram. The positions of the annotations are given by the histogram edges (plus half the bin width). You may then loop over all bins and place a text in each bin.

import matplotlib.pyplot as plt
import numpy as np; np.random.seed(1)

x = np.random.poisson(size=(160))
y = np.random.poisson(size=(160))

fig, ax = plt.subplots()
ax.set_aspect("equal")
hist, xbins, ybins, im = ax.hist2d(x,y, bins=range(6))

for i in range(len(ybins)-1):
    for j in range(len(xbins)-1):
        ax.text(xbins[j]+0.5,ybins[i]+0.5, hist.T[i,j], 
                color="w", ha="center", va="center", fontweight="bold")

plt.show()

If only a single annotation is required, e.g. the following

ax.text(xbins[1]+0.5,ybins[1]+0.5, "HA", 
        color="w", ha="center", va="center", fontweight="bold")

will produce