A Python based 2D random walk simulation tool box

• Allows any type of 2D domain out of different geometrical features
• Allows for any type of random walker, since random walks can be stacked upon each other as superpositions,such as
  • Brownian Walker
  • Correlated Random Walker
  • Composite Random Walker
  • Correlated Composite Random Walker
• Each random walker can respond to each geometrical feature differently with various implemted boundary conditions
• Each random walker can respond to any other random walker differently with various implemted hitting conditions
• Live plotting
• Final Statistics

If you have Python2.7 already installed, simply obtain PyRW

git clone https://github.com/alexblaessle/PyRW

and install via:

python setup.py install

PyRW only depends on

• numpy>=1.8.2
• matplotlib>=1.4.3
• pickle

and a bunch of standard Python2.7 libraries such as

• time
• platform
• sys

The API of pyrw can be found here .

The Documentation of pyrw can be found here .

	import pyrw

	#Create RW
	rw=pyrw.RWRW.RW()

	#Grab domain
	d=rw.domain

	#Build circular domain
	vcenter=d.addVertex([0,0])
	c=d.addCircle(vcenter,35)

	##Add run
	r=rw.addNRuns(1)

	#Draw domain
	d.draw(ann=False)

	#Add predator random walker to the middle of the circular domain
	w=rw.addWalker(BCtyp='setBack',typ='pred',HCtyp='stop',RWtyp='CCRW')

	#Add 5 prey random walkers randomly across the domain
	for i in range(5):
		w2=rw.addWalker(BCtyp='setBack',color='m',x0=[-5,1],typ='prey')
		w2.genRandomX0()

	#Define walker of interest
	rw.setVarForAllRuns('walkerOfInterest',w)

	#Run RW
	rw.runs[0].start(plotStep=True)

	#Print Final statistic
	rw.computeStatistics()
	rw.printStatistics()

	raw_input("Done, press [ENTER] to quit")