A simple python program to estimate STEEM POWER distribution of a g...

Hi friends. This time I have a toy python code for you guys to play around with. This is my first attempt at data analysis with steemit data. The code I have is actually from this reply. I have put it in a loop to run over through a list of steemit handles. The users were those got mentioned in steemSTEM distilled-36 article. These authors got both @curie and @steemstem votes. There was a total of 171 users. I am giving that list here.

Installation first

My machine is a Linux debian system. I assume that you know to install python. What I did was I downloaded python-3.6.5 from here. Extracted and installed it locally. I installed pip package too. (In case any of you want help regarding these, please reply to this article.) Then I installed steem-python as below:
python -m pip install -U git+git://github.com/Netherdrake/steem-python

Python Code

from steem import Steem
s = Steem()
sp_per_mv = 490.355 
#490.355 April 4 (3.47 PM Indian time GMT+5.30)

users = []#steemit usernames
SP = []#steem power
i=0#counter

#users.txt from steemSTEM distilled-36

for line in open('/home/devanandt/Documents/STEEMIT/STEEM-PYTHON/Test_projects/distribution_of_n_users/users.txt'):
    users.append(line.strip('\n')) #appending list with usernames deleting newlines

# the computation of SP
for user in users:
    account = user
    print(user)
    #code-snippet from @kryzsec
    vs = float(str(s.get_account(account)['vesting_shares']).replace(' VESTS', ''))
    rvs = float(str(s.get_account(account)['received_vesting_shares']).replace(' VESTS', ''))
    dvs = float(str(s.get_account(account)['delegated_vesting_shares']).replace(' VESTS', ''))
    SP.append((vs+rvs-dvs)*sp_per_mv/1000000)

#the data writing part
thefile = open('SP.dat', 'w')
for item in SP:
    i=i+1
    thefile.write("%d %s\n" %(i,item))

thefile.close()

This is not an efficient code I am sure. If there are good python coders, you can reply the better versions of it. This is very slow too. Ultimately my dream is to get all steemit user handles and check the wealth distribution and see it the results math with this paper by Dr. Venkat et al. The paper says in a free market economy the wealth distribution tends towards log-normal distribution. They use Statistical mechanics and Game theory concepts to deduce it.

A better Code [from @kryzsec ; See his comment for this article]

from steem import Steem
s = Steem()
sp_per_mv = 490.355 
#490.355 April 4 (3.47 PM Indian time GMT+5.30)

users = []#steemit usernames
SP = []#steem power
i=0#counter

#users.txt from steemSTEM distilled-36

for line in open('./users.txt'):
    users.append(line.strip('\n')) #appending list with usernames deleting newlines



for user in users:
    usr_dict=s.get_account(user)
    print(user)
    #code-snippet from @kryzsec
    vs = float(str(usr_dict['vesting_shares']).replace(' VESTS', ''))
    rvs = float(str(usr_dict['received_vesting_shares']).replace(' VESTS', ''))
    dvs = float(str(usr_dict['delegated_vesting_shares']).replace(' VESTS', ''))
    SP.append((vs+rvs-dvs)*sp_per_mv/1000000)


thefile = open('SP.dat', 'w')
for item in SP:
    i=i+1
    thefile.write("%d %s\n" %(i,item))

thefile.close()

Where do you get STEEM per MV (steem per mega vests)?

This quantity is needed to calculate SP of the user. I got it from steemd.com page. Image below:

Gnuplot code (in semilog style for X axis)

##
reset
set terminal epslatex color standalone#or png
#set style fill  transparent solid 0.5
set key autotitle columnhead
set o "SP.tex"

Min = 0 # where binning starts
Max = 40000 # where binning ends
n = 1000 # the number of bins
binwidth = (Max-Min)/n # binwidth; evaluates to 1.0
bin(x) = binwidth*(floor((x-Min)/binwidth)+0.5) + Min

set xlabel "STEEM POWER"
set ylabel "DISTRIBUTION" offset 2,0
set logscale x
plot "./SP.dat" using (bin($2)):(1.0) smooth freq with lines lt -1 lc 1 title "";

unset output    
set output # finish the current output file
system('latex SP.tex && dvips SP.dvi && ps2pdf SP.ps')
system('mv SP.ps SP.eps')
unset terminal
reset

Semilog Plot

Raw plot (with no semilog command)

Barely seeing anything right! :P

Zooming into Raw plot (0 to 500 SP)

Conclusion

The only conclusion as I understand is most of the wealth at least in terms of SP is in few hands. Let us initiate a discussion about the science of economics too.

Acknowledgement

I thank @kryzsec for this detailed reply with a coding example. He pointed me to a very nice article written by @themarkymark for installing the library steem-python (which works with python3 only).

Reference

Software used

Python 3.6.5 (with steem-python library for data pulling)
Gnuplot (for plotting)

A call to users who may be interested in these kinds of data analysis

Anyone interested in similar kind of data analysis? Let me know.
@tibra : Maybe you want to see this article :P
@bharathchand : What do you think?

If you like this post, please upvote and resteem it.

Below are my recent posts, you may find it interesting:

Join #steemSTEM

Join the active science community #steemSTEM at discord: https://discord.gg/BZXkmWw

And to steemSTEM beginners:

You can ask for help in our discord page. There are people ready to help you there.

gif courtesy: @rocking-dave

All images without image sources are my creations :)

Follow me @dexterdev

 ____ _______  ______ _________ ____ ______    
/  _ /  __\  \//__ __/  __/  __/  _ /  __/ \ |\
| | \|  \  \  /  / \ |  \ |  \/| | \|  \ | | //
| |_/|  /_ /  \  | | |  /_|    | |_/|  /_| \// 
\____\____/__/\\ \_/ \____\_/\_\____\____\__/

A simple python program to estimate STEEM POWER distribution of a group of steemSTEM authors[Updated]