Tracking History

The tools package contains utilities that interact with the learning process itself. For example, tools.lineage has utilities that enable tracking the lineage of individuals.

These utilities are as follows:

• TrackParents modifies the behaviour of an existing variator to record the parents in each offspring.

• graph_lineage takes a collection of individuals and plot their lineage. These individuals must be produced by variators enhanced with TrackParents.

import random
random.seed(1)
from evokit.core import Population
from evokit.evolvables.bitstring import BitString, CountBits
from evokit.tools.lineage import TrackParents
from evokit.tools.lineage import graph_lineage

To enable lineage tracking, take an existing algorithm (e.g. OneMax) and wrap the variator with TrackParents. Then, run the algorithm as normal.

T: int = 10

from evokit.evolvables.algorithms import SimpleLinearAlgorithm
from evokit.evolvables.bitstring import OnePointCrossover
from evokit.evolvables.selectors import TruncationSelector

another_pop = Population((BitString.random(size=20)
                                      for _ in range(50)))

ctrl = SimpleLinearAlgorithm(another_pop,
                             CountBits(),
                             TruncationSelector(budget=6),
                             TrackParents(OnePointCrossover(0.5), 9))

for _ in range (10):
    ctrl.step()

Accessing Parents

TrackParents stores the parents of each offspring in its .parents. This can incur a memory cost: if an individual has at most parents and generations of parents are tracked, then at most

extra individuals may be stored in memory. The cost can be significantly less in practice, as in the following example, because successful individuals often share common ancestors.

ctrl.population[0].parents

(<evokit.evolvables.bitstring.BitString at 0x1f30dc14a70>,
 <evokit.evolvables.bitstring.BitString at 0x1f30dc15130>)

Visualising Parents

The graph_lineage function takes a number of individuals and outputs a Digraph (type from the graphviz module).

Connections can look messy with large populations. In this case, either reduce the node size with use_tooltip=True or increase vertical spacing with vertical_spacing=N, N>1.

Note that tooltips are only visible when the file is saved in SVG format. Do this with by setting save_as=<file_name>.

dot = graph_lineage(ctrl.population, use_tooltip=True, save_as="weeeeee")
dot