Progress Bar For Multi Processing in Python

progbar.py module that I created is for monitoring the progress of the processes. Best competible with Monte Carlo (MC) simulations.

Competible Multiporcessing functions:

* Pool.map()
* Pool.imap()
* Pool.map_async()
* Process

This piece of algorithm pops up a proggress bar on to your terminal for multiprocessing (can be used with pool.map, pool.imap, pool.map_async, and process) which contains id of your processes, estimated execution time, launch time, percentage of the work done etc.) Here is the recipe for using it in a proper way!

Recipe

What you need to do is:

Download progbar.py and be sure that this file downloaded to the same directory with your __main__ code.
Import time, os, inspect into your main code to let progbar module stamps the versions of Scipy, Numpy, GCC, current file name and directory and let module estimates the execution time of the code, i.e.

import time
import sys
import inspect, os

Then, import progbar.py into your main code.

from progbar import *

Call stamp_versions() for letting you know the versions of Scipy, Numpy, GCC, current file name and directory :

Lastly you should embed progress_bar(pool_count), progress_estimator(i,current_xcoordinate,Nrun1), percentage(current_pid, i, current_xcoordinate, Nrun1,launch_time,c_launch_time) correctly to your MC sampler function : For example let our MC sampler function be run(Nrun1), Then we have,

def run(Nrun1):
	
	global launch_time,c_launch_time
	progress_bar(pool_count)
	#import the times when the code first launched
	launch_time=time.time()
	c_launch_time=time.ctime()
	
	#multiprocessing in queue (to use other module comment the line below)
	q.put([Nrun1])
	
	for i in range (Nrun1):
	
		#your current process:
		current = multiprocessing.current_process()
		#current coordinate of your cursor
		current_xcoordinate=current._identity[0]
		#current process id
		current_pid=os.getpid()
		#puts '>' when some Nrun1/25 of the job completed
		progress_estimator(i,current_xcoordinate,Nrun1)
		#pops up the percentage, launch date of the code (including year month and day),
		#estimated execution period of the code (your computer will do the job in 'X.XXX'ours)
		percentage(current_pid, i, current_xcoordinate, Nrun1,launch_time,c_launch_time)
		
		#do some complicated work
		time.sleep(0.004)

For a complete example here is the link.

Acknowlegement

Finally, I would like to thank Burak Kakillioğlu for enforcing me to upload this module on GitHub.

^{This work is dedicated to HÖ.}

bkakilli/prog-bar

Progress Bar For Multi Processing in Python

Recipe

Acknowlegement