spread values of dataset equally over fixed number of bins

[moose_man]

Hi all,

I used numpy to bin some data with following simple code:

a = np.array(data)
histo = np.histogram(a,bins=nbrbins)

But the results look like this (left = bin, right = number of samples)
0.0->11356
5.7143->68
11.4286->18
17.1429->8
22.8572->3
28.5715->2
34.2858->1
40.0001->1
45.7144->6
51.4287->3

I would like to have all values more spread, so that the 0.0-5.7 bin would be split into smaller bins.
Which approach are you guys taking here? Should I look into other statistical packages?
In fact I would need bins that vary in size, but I still need the same number of bins. As far as I see, this couldn't be done with numpy.

Rgds,
M.

[Gribouillis]

You have many options for the 'bins' argument. They are described in the docstring of numpy.histogram()

[moose_man]

yes indeed, but none of the options worked. I want to have a fixed number of bins , and that's the problem. The options change the number of bins.

[ichabod801]

You can set the bins option to an array of the bin edges, and have whatever bins you want.