Using array to generate random text

Question

I am trying to generate random text using letter frequencies that I have obtained. First, I succeeded with the following code:

for i in range(450):
    outcome=random.random()
    if 0<outcome<0.06775:
        sys.stdout.write('a')
    if 0.06775<outcome<0.07920:
        sys.stdout.write('b')
    if 0.07920<outcome<0.098:
        sys.stdout.write('c')
    ....

This until the letter z and spacebar. This give me >50 lines of code and I want to get the same result using an array.

So far I have :

f_list = [0, 0.06775, 0.08242, 0.10199, 0.13522, 0.23703, 0.25514, 0.27324, 0.32793, 0.38483, 0.38577, 0.39278, 0.42999, 0.45023, 0.50728, 0.56756, 0.58256, 0.58391, 0.62924, 0.68509, 0.7616, 0.78481, 0.79229, 0.81161, 0.81251, 0.82718, 0.82773, 0.99998]
alphabet = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', ' ']

import random
import sys

for i in range(25):
    outcome=random.random()
    if f_list[i]<outcome<f_list[i+1]:
        sys.stdout.write('alphabet[i]')

But it isn't working properly, as the range seems now to relate to the array and not the number of iterations I want. The output is blank.

(the frequencies in the 2nd code are the correct ones)

Julia
– Julia

2011-12-21 11:48:14 +00:00
Commented Dec 21, 2011 at 11:48 — Julia
– Julia, Commented Dec 21, 2011 at 11:48

eumiro · Accepted Answer · 2011-12-21 11:58:30Z

17

import random
import sys
import bisect

f_list = [0, 0.06775, 0.08242, 0.10199, 0.13522, 0.23703, 0.25514, 0.27324, 0.32793, 0.38483, 0.38577, 0.39278, 0.42999, 0.45023, 0.50728, 0.56756, 0.58256, 0.58391, 0.62924, 0.68509, 0.7616, 0.78481, 0.79229, 0.81161, 0.81251, 0.82718, 0.82773, 0.99998]
alphabet = 'abcdefghijklmnopqrstuvwxyz '

for i in xrange(450):
    sys.stdout.write(alphabet[bisect.bisect(f_list, random.random()) - 1])

does the trick and returns (example):

l wefboethol gsplotfoh ua onpedefh dnolnairnioeiegehhecaworonnfmeuej dsiauhpbfttwcknal ateof ap cgbr sunnee leseaeeecltaiaur u oen vxntgsoio kdeniei ot df htr dcencrsrrfp bwelsuoaslrnr heh ee tpt oeejaldeatcto fi a u idimiadmgglral o m iaielbtnt es oe shlspudwdfrrsvol oo i tlwh d r i swhsnloai p swlooi wbe nn sshth nsawtnrqsud mtw diit pner r nitmah todf zcsehma hl e ros ctee toiouinn i hl hlonphioe nh gan ho heein itrgeylftn epaacrmanhe

alphabet can be defined as a simple string too (accessing its elements - single characters - works like for lists)

bisect.bisect(list, value) takes a sorted list and a value and tells where this value should be put between. More about bisect.

edited Dec 21, 2011 at 11:58

answered Dec 21, 2011 at 11:50

eumiro

214k36 gold badges307 silver badges264 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

v1Axvw Over a year ago

+1 for the binary search algorithm usage. People are still using to much algorithms in linear time.

DhruvPathak Over a year ago

@eumiro need your help to understand the binary search part.

eumiro Over a year ago

@DhruvPathak - you can find some examples on the linked documentation page.

Julia Over a year ago

@eumiro How do I store this random text into a name, say, "text1", in order to use it later in my program? Thank you!

Don Question · Accepted Answer · 2011-12-21 14:16:41Z

Eumiros answer is perfect, and much simpler then mine, but because i made the effort to modify a older solution to a similar problem, i don't want it to go to waste.

I even had the link still around for the discussion about weighted random generators from which i borrowed the "King of the hill" algorithm.

from string import lowercase
from random import random

class TextGenerator(object):        
        def __init__(self, flist, textlength, charmap = lowercase + ' '):            
            self.text_length = textlength
            self.chars = charmap
            self.weights = self._get_weight_list(flist)            

        def create_new_weights(self, flist):
            self.weights = self._get_weight_list(flist)

        def get_weight(self, char):
            return self.weights[self.chars.index(char)]            

        def change_weight(self, char, weight):
            self.weights[self.chars.index(char)] = weight

        def _get_weight_list(self, flist):
            return map (lambda x, y: y-x,
                        flist,
                        flist[1:] + [1.0])[:-1]

        def windex(self):
            assert(len(self.weights) == len(self.chars))
            rnd = random() * sum(self.weights)
            for i, w in enumerate(self.weights):
                rnd -= w
                if rnd < 0:
                    return i

        def create_text(self, flist = None):
            weights = self._get_weight_list(flist)if flist else self.weights
            return u''.join([self.chars[self.windex()] for i in range(self.text_length)])

flist = [0, 0.067750000000000005, 0.082419999999999993, 0.10199, 0.13522000000000001, 0.23702999999999999, 0.25513999999999998, 0.27323999999999998, 0.32793, 0.38483000000000001, 0.38577, 0.39278000000000002, 0.42998999999999998, 0.45023000000000002, 0.50727999999999995, 0.56755999999999995, 0.58255999999999997, 0.58391000000000004, 0.62924000000000002, 0.68508999999999998, 0.76160000000000005, 0.78481000000000001, 0.79229000000000005, 0.81161000000000005, 0.81250999999999995, 0.82718000000000003, 0.82772999999999997, 0.99997999999999998]

texter = TextGenerator(flist, 1000)
print texter.create_text()

texter.change_weight('i', texter.get_weight('e') * 2)
print texter.create_text()

Collectives™ on Stack Overflow

Using array to generate random text

2 Answers 2

4 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

4 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related