python-hello-evolution

To run type

python hello.py

To run with custom parameters

python hello.py -t MyHelloString -p 100

where

• -p is number of population (default is 10000)
• -t is looked string (default Hello-Evolution)

Sample output

Generation | Fitness | Phenotype
--------------------------------------
         1 |   148.0 | ?itti*Bb}tWzl`y
         2 |   122.0 | ?Xpm|4Hs]{v~Xpu
         3 |    93.0 | Ha]gr2<stcvlQpo
         4 |    63.0 | =ii_t.Nvoiswgjl
         5 |    46.0 | Iamtt.Nvoiswgjl
         6 |    33.0 | Cfhon.Fpplypinm
         7 |    26.0 | Helgn(Hvimwtinl
         8 |    23.0 | Ielmo.Broiswgnl
         9 |    16.0 | Helqp*Cvonvtinm
        10 |    10.0 | Helmo.Bvpiutioo
        11 |     6.0 | Helmo.Gvomutinn
        12 |     4.0 | Helmo.Fvolutioo
        13 |     2.0 | Hello.Dvolution
        14 |     2.0 | Hello.Dvolution
        15 |     0.0 | Hello-Evolution

How it works?

It uses evolution algorithm to solve desired string.

In short it generates about 10 000 random strings (as population). These strings are filtered by fitness method to 20% of all population.

Fitness method checks how close string is close to desired string.

Crossover mixes randomly string and fill the population of new the string. Crossover method is repeated until population length will get initial length.

Mutation does a little change in string for example changes random character.

These steps are repeated until we get desired string.

To run test type

py.test tests.py