Logarithmic Binning Histogram

[Noa_A]

I want to plot the logarithmic binning histogram of a 1D array in LabView.

It means the x-scale is logarithmic and the bins are equally spaced in log10.

I have done it in python and I'm using this Python function in LabView to plot logarithmic binning. But it is not efficient and I want to do it in LabView.

Python function:

import numpy as np
import matplotlib.pyplot as plt

def f(data_array):
      mini=min(data_array)
      maxi=max(data_array)

      bins=np.logspace(np.log10(mini), np.log10(maxi), 80)
      widths=(bins[1:] - bins[:-1])

      n, bins, patches=plt.hist(data_array, bins, density=True)

      plt.xscale('log')
      plt.xlabel('interval times(s)')
      plt.ylabel('number')
      return plt.show()

My LabView code:

Do you have any idea how can I do that?

I can plot a histogram in LabView with Logarithmic x-scale but the bins are not logarithmic:

[kartiknattar]

 go to graph properties. select scales tab then select X-Axis and enable log

[Noa_A]

Kartiknattar,

Thanks for your response, I already have done it. 

As I mentioned I need the bins to be equally spaced in log10.

If I choose the interval a constant number and just choose the x-scale logarithmic, then I will have this:

This is not what I want. As you see the bins are not equal in the log scale.

[kartiknattar]

Try this example code. It might help you..

https://forums.ni.com/t5/Example-Code/Programmatically-Set-A-Graph-Scale-To-Linear-Or-Log/ta-p/35244...

[Noa_A]

For example, by this simple code and choosing x-scale logarithmic but interval as a constant number:

I will have:

But with the Python code, for the same array I have:

That is what I want. So the x scale is not the problem, I should do some thing with the interval. but I dont know how. 

In python I have set the bins:

   bins=np.logspace(np.log10(mini), np.log10(maxi), 80)

so, the intervals of the bins are Logarithmic. 

[Noa_A]

Actually, my Python code had some problems 😁, The correct one is this:

   data=[10, 10, 100, 2, 2, 10, 10, 1, 1, 100, 10, 10, 50, 50, 50, 50, 100, 100, 100]
   maxi=max(data)
   mini=min(data)

   pl.hist(data, bins=np.logspace(np.log10(mini),np.log10(maxi), 80))
   pl.gca().set_xscale("log")
   pl.show()

The result is:

So, I expect this result from LabView.

[James_W]

---

@Noa_A wrote:

Actually, my Python code had some problems 😁, The correct one is this:

   data=[10, 10, 100, 2, 2, 10, 10, 1, 1, 100, 10, 10, 50, 50, 50, 50, 100, 100, 100]
   maxi=max(data)
   mini=min(data)

   pl.hist(data, bins=np.logspace(np.log10(mini),np.log10(maxi), 80))
   pl.gca().set_xscale("log")
   pl.show()

The result is:

So, I expect this result from LabView.

---

Have we missed something? Your "Python code had some problems" so you assume the issue is with LabVIEW not Python?

[Noa_A]

The problem is that I dont want to use Python. I just want to have the bins that are equally spaced in the x scale (that is logarithmic) all in labview.

[Noa_A]

I mean how can I define "np.logspace" in LabView to wire it to the interval of LabView histogram.

[James_W]

---

@Noa_A wrote:

Kartiknattar,

Thanks for your response, I already have done it. 

As I mentioned I need the bins to be equally spaced in log10.

If I choose the interval a constant number and just choose the x-scale logarithmic, then I will have this:

This is not what I want. As you see the bins are not equal in the log scale.

---

Are you sure you know what you want?
The bins above are equally spaced in the log scale, as the scale goes up, so the bins get closer - and bin width and distance between the bins halves (the definition of a log scale).
Equally spaced (the same width) would not be a log scale!