python - imshow and histogram2d: can't get them to work

Question

I'm learning Python and this is my first question here. I've read other topics related to the usage of imshow but didn't find anything useful. Sorry for my bad English.

I have plotted a set of points here, left graphic:

points (left) and image (right)

Now I'd like to see an image of the density of points, so I used imshow and histogram2d, and I got the image to the right in the previous link.

The image doesn't correspond to the distribution of points. How is this possible? I've followed the instructions in the help and even changed some parameters but nothing worked :(

The code is:

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.cm as cm

j, h, k = np.loadtxt("test.dat", usecols=(2, 4, 6), \
    unpack=True)

# límites
xmin = -0.5
xmax =  3.0
ymin = -0.5
ymax =  4.0

# colores
j_h = j - h
h_k = h - k

# no todas las estrellas son graficadas    
x1 = 0.5
y1 = 0.5
b  = 2.2
c  = y1 - b * x1

x = y = np.array([])

for xi, yi in zip(h_k, j_h):
    if xi < (yi - c) / b:
        x = np.append(x, xi)
        y = np.append(y, yi)

# gráfico
fig = plt.figure(figsize=(8, 7))

ax = fig.add_subplot(111)
#ax.plot(x, y, "go")
ax.set_xlabel(r"X", fontsize=14)
ax.set_ylabel(r"Y", fontsize=14)
ax.axis([xmin, xmax, ymin, ymax])

# imagen
rango = [[xmin, xmax], [ymin, ymax]]
binsx = int((xmax - xmin) / 0.05)
binsy = int((ymax - ymin) / 0.05)
binsxy = [binsx, binsy]

H, xedges, yedges = np.histogram2d(x, y, range=rango, bins=binsxy)

extent = [yedges[0], yedges[-1], xedges[0], xedges[-1]]
cp = ax.imshow(H, interpolation='bilinear', extent=extent, cmap=cm.jet)
fig.colorbar(cp)

plt.show()

The links for the data used is here:

https://dl.dropbox.com/u/10411539/python/test.dat

Any help is appreciated!

Answer 1

Try different interpolation, and transpose the matrix to get it in the same axis:

cp = ax.imshow(H.transpose()[::-1], interpolation='nearest', extent=extent, cmap=cm.jet)

Answer 2

Is this what you want to get? You can use pcolor (and pcolormesh) if you want to pass the x and y coordinates.

import urllib
import numpy as np
import matplotlib.pyplot as plt
f = urllib.urlopen('https://dl.dropbox.com/u/10411539/python/test.dat')
j, h, k = np.loadtxt(f, usecols=(2, 4, 6), \
    unpack=True)
j, h, k
j_h = j - h
h_k = h - k
H, xedges, yedges = np.histogram2d(j_h, h_k, bins=100)
plt.pcolor(xedges, yedges, H)

For imshow you have to reverse the first dimension, because imshow uses zero-based row, column indices to the x, y. Drawing from the top down.

plt.imshow(H[::-1,:], extent=(0,5, 0,2.5)) # typed in extent by hand.