AJW's site

Using a command line parameter to specify the name of a function to call

When running some Python experiments — where typically each experiment corresponds to run a simulation under some particular conditions — I find useful to be able to call a particular function from the command line. For example, if I define the functions exp_1, exp_2, and exp_3, I want to be able to do the following from the command line:

$ python my_script.py exp_1
Running experiment 1...

The following code allows this:

def exp_1():
    print 'Running experiment 1...'
    # Do some stuff

def exp_2():
    print 'Running experiment 2...'
    # Do some stuff

def exp_3():
    print 'Running experiment 3...'
    # Do some stuff

if __name__ == '__main__':
    import sys

    if len(sys.argv) == 2:
        fname = sys.argv[1]
        if locals().has_key(fname) and hasattr(locals()[fname], '__call__'):
            locals()[fname]()
        else:
            print 'Function %s does not exist' % fname
    else:
        print 'Wrong number of arguments'

Note that if you're using IPython, you can do

In [1]: run foo exp_1
Running experiment 1...

Older articles

Binomial Coefficients in NumPy/SciPy
Sun 21 October 2012
By Alejandro Weinstein

In python.

For some reason finding how to compute the binomial coefficients using Numpy/SciPy is harder than it should (I couldn't find the answer in the first page of a google search).

The answer, though, is quite easy: just use scipy.special.binom. For example:
```
>> import scipy
>> print scipy.special ...
```
read more
Using NumPy masked arrays

Tue 02 October 2012
By Alejandro Weinstein

In python.

In NumPy, masked arrays are quite handy. Say you want to find the index of the smallest positive entry of an array. Doing a[a>0].argmin() doesn't work, because you're removing all the non-positive numbers from a, so you're computing argmin over the wrong array.

Masked ...
read more

Page 1 / 1

Older articles

Follow