2d_histogram.py

"""
hexbin is an axes method or pyplot function that is essentially
a pcolor of a 2-D histogram with hexagonal cells.  It can be
much more informative than a scatter plot; in the first subplot
below, try substituting 'scatter' for 'hexbin'.
"""

import numpy as np
import matplotlib.cm as cm
import  matplotlib.pyplot as plt
from matplotlib import mlab as ML

filename = './temporal_series/most_weigh_ins/number_weigh_ins_vs_time_span.dat'
a_file = open(filename,'r')
data = a_file.readlines()
a_file.close()


dist = []
prob = []
deltan = []
for row in data:
    row = row.split()
    
    dist.append(row[0])  # x column
    prob.append(row[1])  # y column


    prob_ij = map(lambda x: float(x), prob)
    dist_ij = map(lambda x: float(x), dist)
   
xmin = min(dist_ij)  #or  i can pick the limits manually here
xmax = 1000.0#max(dist_ij)
ymin = min(prob_ij)
ymax = max(prob_ij)


gridsize=40

print xmin ,xmax, ymin, ymax


plt.ylabel ('# weigh-ins', size = 20)
plt.xlabel('timespan (days)', size = 20)

plt.hexbin(dist_ij,prob_ij, bins='log',gridsize=gridsize, cmap=cm.jet,mincnt=1)   #hexagonal bins.  
#if bins=None, the color of each hexagon directly corresponds to its count value
#if bins=integer, divide the couts in the specified number of bins
#if bins='log' use a logarithmic color scale

#gridsize is the NUMBER of hexagons in the x-direction (=100 by default)
#mincnt=1 to diferenciate in colors between count=0 and count=1

ax = plt.gca()
fontsize = 18
for tick in ax.xaxis.get_major_ticks():
    tick.label1.set_fontsize(fontsize)
for tick in ax.yaxis.get_major_ticks():
    tick.label1.set_fontsize(fontsize)


#ax.set_xscale('log')
#ax.set_yscale('log')


plt.axis([xmin, xmax, ymin, ymax])
#plt.title("With a log color scale")
cb = plt.colorbar()
cb.set_label('log_10(N+1)', size = 18)

plt.show()