How to show numpy NxM array with dtype=float as plain gray scale image?

Question

When creating an numpy array with dtype=float, the the presentation method using matplotlib.pyplot.imshow appears to be dependent on the v

Answer 1

Found out that imshow does normalize by default, so to avoid this, the vmin and vmax must be given to imshow, which is then:

plt.imshow(img, cmap='gray', vmin=0, vmax=1)

Answer 2

You have to fix the limits of the color-scale:

plt.imshow(img, cmap='gray',clim=(0,1))

To get a good feeling of what is going on you could include a colorbar which visualizes the conversion between colors and numerical values; for example using the following code:

fig,ax = plt.subplots()
cax = plt.imshow(img, cmap='gray')
cbar = fig.colorbar(cax)
plt.show()

Doing this for the two examples immediately makes clear that matplotlib.pyplot updates the range of the color-scale to the data. Consequently the conversion betweens colors and numerical values is different for the two cases.