Command line argument in Histogram

Question

I am writing some code which makes a histogram out of imported excel files from a directory and bins the data according to some parameter and exports new excel files accordingly, binned into their respective bins, for ex. if there is a number 5.16 then the bin count for bin (5,10] will go up by one and so on. However, I wanted to write something in which I can input a specific value that would change the bins accordingly, like if I wanted bins of 3, I would choose n=3 and the code would now bin accordingly, such that it makes bin (0,3], (3,6], etc. and the same rules as previously would apply. The original code is:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import openpyxl
from pandas import ExcelWriter
import os

datadir = '/Users/user/Desktop/Newfolder/'

for file in os.listdir(datadir):
    if file.endswith('.xlsx'):
        data = pd.read_excel(os.path.join(datadir, file))
        counts, bins, patches = plt.hist(data.values, bins=range(0, 
        int(max(data.values)+5), 5))
        df = pd.DataFrame({'bin_leftedge': bins[:-1], 'count': counts})
        plt.title('Data')
        plt.xlabel('Neuron')
        plt.ylabel('# of Spikes')
        plt.show()

        outfile = os.path.join(datadir, file.replace('.xlsx', '_bins.xlsx'))
        writer = pd.ExcelWriter(outfile)
        df.to_excel(writer)
        writer.save()

My thought was to use argparse as a command line argument which I could us to input the change in the bins, so I wrote:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import openpyxl
from pandas import ExcelWriter
import os
import argparse

datadir = '/Users/user/Desktop/Newfolder/'

parser = argparse.ArgumentParser(description = 'Calculating the bin width')
parser.add_argument('n', type=int, help='Changing of the width')
args = parser.parse_args()

def vary(n):
    wid = n
     return wid

if __name__ == '__main__':
    print(vary(args.n))

for file in os.listdir(datadir):
    if file.endswith('.xlsx'):
       data = pd.read_excel(os.path.join(datadir, file))
       counts, bins, patches = plt.hist(data.values, bins=range(0, 
       int(max(data.values)+vary(n)), vary(n)))
       df = pd.DataFrame({'bin_leftedge': bins[:-1], 'count': counts})
       plt.title('Data')
       plt.xlabel('Neuron')
       plt.ylabel('# of Spikes')
       plt.show()

       outfile = os.path.join(datadir, file.replace('.xlsx', '_bins.xlsx'))
       writer = pd.ExcelWriter(outfile)
       df.to_excel(writer)
       writer.save()

I apologize in advance if this looks idioitic, since I am pretty new to coding and still don't know too much about anything. Anyway, I get an error say that

Traceback (most recent call last):
File "\Users\user\Desktop\Bins.py", line 25, in <module>
        counts, bins, patches = plt.hist(data.values, bins=range(0, int(max(data.values)+vary(n)), vary(n)))
NameError: name 'n' is not defined

Could I get some help with this, how would I go about implementing this command line argument ( argparse ) into the histogram so I can input the according bins everytime I need to change them. Any help would be greatly appreciated, thank you

Answer 1

Without going too much into your code, command line arguments are easily processes using the sys module:

import sys
print sys.argv # first element is the script name, then follow the parameters as strings

So that this script, if I name it sysArgs.py and call it with some parameters in the console, prints

python sysArgs.py lala 5
['sysArgs.py', 'lala', '5']

If you want to only pass one parameter n , then change it to

import sys
n = int(sys.argv[1])
# do stuff with n other than printing it
print n

Answer 2

Here's how I'd organize your code. main can be reorganized, but my focus is on the location of the commandline parsing.

datadir = '/Users/user/Desktop/Newfolder/'
n = 3  # a default if not used with argparse

def main(datadir, n):
    # might split the load and the plot functions
    # or put the action for one file in a separate function

    for file in os.listdir(datadir):
        if file.endswith('.xlsx'):
           data = pd.read_excel(os.path.join(datadir, file))
           counts, bins, patches = plt.hist(data.values, bins=range(0, 
           int(max(data.values)+n), n))
           df = pd.DataFrame({'bin_leftedge': bins[:-1], 'count': counts})
           plt.title('Data')
           plt.xlabel('Neuron')
           plt.ylabel('# of Spikes')
           plt.show()

           outfile = os.path.join(datadir, file.replace('.xlsx', '_bins.xlsx'))
           writer = pd.ExcelWriter(outfile)
           df.to_excel(writer)
           writer.save()

if __name__ == '__main__':
    # run only as a script; not on import
    # if more complicated define this parser in a function

    parser = argparse.ArgumentParser(description = 'Calculating the bin width')
    parser.add_argument('n', type=int, help='Changing of the width')
    args = parser.parse_args()

    print(args)  # to see what argparse does
    main(datadir, args.n)
    # main(args.datadir, args.n) # if parser has a datadir  argument

(I haven't tested this.)