Ignore:
Timestamp:
07/05/24 02:08:57 (3 months ago)
Author:
Maciej Komosinski
Message:

Cosmetic

File:
1 edited

Legend:

Unmodified
Added
Removed
  • framspy/FramsticksLib.py

    r1310 r1312  
    3636        DETERMINISTIC: bool = False  # set to True to have the same results in each run
    3737
    38         GENOTYPE_INVALID = "/*invalid*/"  # this is how genotype invalidity is represented in Framsticks (Geno.format is 'invalid'). Mutation and crossover operators return such a genotype if they were unable to perform their operation (information about the cause is stored in the Geno.info field - see GenMan.cpp)
    39         GENOTYPE_INVALID_OFFSPRING_SUBSTITUTE_ORIGINAL = True  # if True, when mutation or crossover is unable to perform their operation for the provided genotype(s), return the original genotype (and print a warning). If this happens extremely rarely, it may be ignored - but if not, you need to identify the reason (e.g., particular genotypes that cause the problem), fix it or change the logic of your algorithm. A more strict approach is to keep this field False - then you must always check if GENOTYPE_INVALID was returned by mutate() or crossOver(), and handle this situation properly (e.g., choose different parent(s) for mutate() or crossOver() and repeat until you get a valid offspring).
     38        GENOTYPE_INVALID: str = "/*invalid*/"  # this is how genotype invalidity is represented in Framsticks (Geno.format is 'invalid'). Mutation and crossover operators return such a genotype if they were unable to perform their operation (information about the cause is stored in the Geno.info field - see GenMan.cpp)
     39        GENOTYPE_INVALID_OFFSPRING_SUBSTITUTE_ORIGINAL: bool = True  # if True, when mutation or crossover is unable to perform their operation for the provided genotype(s), return the original genotype (and print a warning). If this happens extremely rarely, it may be ignored - but if not, you need to identify the reason (e.g., particular genotypes that cause the problem), fix it or change the logic of your algorithm. A more strict approach is to keep this field False - then you must always check if GENOTYPE_INVALID was returned by mutate() or crossOver(), and handle this situation properly (e.g., choose different parent(s) for mutate() or crossOver() and repeat until you get a valid offspring).
    4040
    4141        EVALUATION_SETTINGS_FILE = [  # all files MUST be compatible with the standard-eval expdef. The order they are loaded in is important!
Note: See TracChangeset for help on using the changeset viewer.