config_file.rst

Configuration file description

The configuration file is in the format described in the Python :py:mod:`configparser` documentation as "a basic configuration file parser language which provides a structure similar to what you would find on Microsoft Windows INI files."

Most settings must be explicitly enumerated in the configuration file. (This makes it less likely that library code changes will result in your project silently using different NEAT settings. There are some defaults, as noted below, and insofar as possible new configuration parameters will default to the existing behavior.)

Note that the Config constructor also requires you to explicitly specify the types that will be used for the NEAT simulation. This, again, is to help avoid silent changes in behavior.

The configuration file is in several sections, of which at least one is required. However, there are no requirements for ordering within these sections, or for ordering of the sections themselves.

[NEAT] section

The NEAT section specifies parameters particular to the generic NEAT algorithm or the experiment itself. This section is always required, and is handled by the Config class itself.

.. index:: ! fitness_criterion

• fitness_criterion

  The function used to compute the termination criterion from the set of genome fitnesses. Allowable values are: min, max, and mean

.. index:: ! fitness_threshold

.. index:: found_solution()

• fitness_threshold

  When the fitness computed by fitness_criterion meets or exceeds this threshold, the evolution process will terminate, with a call to any registered reporting class' :py:meth:`found_solution <reporting.BaseReporter.found_solution>` method.

Note

The found_solution method is not called if the maximum number of generations is reached without the above threshold being passed (if attention is being paid to fitness for termination in the first place - see no_fitness_termination below).

.. index:: ! no_fitness_termination

.. index:: found_solution()

• no_fitness_termination

  If this evaluates to True, then the fitness_criterion and fitness_threshold are ignored for termination; only valid if termination by a maximum number of generations passed to :py:meth:`population.Population.run` is enabled, and the found_solution method is called upon generation number termination. If it evaluates to False, then fitness is used to determine termination. This defaults to "False".

  .. versionadded:: 0.92
  
  

.. index:: ! pop_size

• pop_size

  The number of individuals in each generation.

.. index:: ! reset_on_extinction

• reset_on_extinction

  If this evaluates to True, when all species simultaneously become extinct due to stagnation, a new random population will be created. If False, a CompleteExtinctionException will be thrown.

.. index:: stagnation

.. index:: DefaultStagnation

[DefaultStagnation] section

The DefaultStagnation section specifies parameters for the builtin DefaultStagnation class. This section is only necessary if you specify this class as the stagnation implementation when creating the Config instance; otherwise you need to include whatever configuration (if any) is required for your particular implementation.

.. index:: ! species_fitness_func

• species_fitness_func

  The function used to compute species fitness. This defaults to ``mean``. Allowed values are: max, min, mean, and :py:func:`median <math_util.median>`

Note

This is not used for calculating species fitness for apportioning reproduction (which always uses mean).

.. index:: ! max_stagnation

• max_stagnation

  Species that have not shown improvement in more than this number of generations will be considered stagnant and removed. This defaults to 15.

.. index:: ! species_elitism

• species_elitism

  The number of species that will be protected from stagnation; mainly intended to prevent total extinctions caused by all species becoming stagnant before new species arise. For example, a species_elitism setting of 3 will prevent the 3 species with the highest species fitness from being removed for stagnation regardless of the amount of time they have not shown improvement. This defaults to 0.

.. index:: reproduction

.. index:: DefaultReproduction

[DefaultReproduction] section

The DefaultReproduction section specifies parameters for the builtin DefaultReproduction class. This section is only necessary if you specify this class as the reproduction implementation when creating the Config instance; otherwise you need to include whatever configuration (if any) is required for your particular implementation.

.. index:: ! elitism

• elitism

  The number of most-fit individuals in each species that will be preserved as-is from one generation to the next. This defaults to 0.

.. index:: ! survival_threshold

• survival_threshold

  The fraction for each species allowed to reproduce each generation. This defaults to 0.2.

.. index:: ! min_species_size

• min_species_size

  The minimum number of genomes per species after reproduction. This defaults to 2.

.. index:: genome

.. index:: DefaultGenome

[DefaultSpeciesSet] section

The DefaultSpeciesSet section specifies parameters for the builtin DefaultSpeciesSet class. This section is only necessary if you specify this class as the genome implementation when creating the Config instance; otherwise you need to include whatever configuration (if any) is required for your particular implementation.

.. index:: genomic distance

.. index:: ! compatibility_threshold

.. index:: species

• compatibility_threshold

  Individuals whose :term:`genomic distance` is less than this threshold are considered to be in the same :term:`species`.

[DefaultGenome] section

The DefaultGenome section specifies parameters for the builtin DefaultGenome class. This section is only necessary if you specify this class as the genome implementation when creating the Config instance; otherwise you need to include whatever configuration (if any) is required for your particular implementation.

.. index:: activation function

.. index:: mutation

.. index:: node

.. index:: attributes

.. index:: X_default

• activation_default

  The default :term:`activation function` :term:`attribute <attributes>` :py:meth:`assigned <attributes.StringAttribute.init_value>` to new :term:`nodes <node>`. If none is given, or "random" is specified, one of the activation_options will be chosen at random.

.. index:: mutate_rate

• activation_mutate_rate

  The probability that :term:`mutation` will replace the node's activation function with a :py:meth:`randomly-determined <attributes.StringAttribute.mutate_value>` member of the activation_options. Valid values are in [0.0, 1.0].

.. index:: X_options

• activation_options

  A space-separated list of the activation functions that may be used by nodes. This defaults to :ref:`sigmoid <sigmoid-label>`. The built-in available functions can be found in :ref:`activation-functions-label`; more can be added as described in :ref:`customization-label`.

.. index:: aggregation function

.. index:: mutation

.. index:: node

.. index:: attributes

.. index:: X_default

• aggregation_default

  The default :term:`aggregation function` :term:`attribute <attributes>` :py:meth:`assigned <attributes.StringAttribute.init_value>` to new :term:`nodes <node>`. If none is given, or "random" is specified, one of the aggregation_options will be chosen at random.

.. index:: mutate_rate

• aggregation_mutate_rate

  The probability that :term:`mutation` will replace the node's aggregation function with a :py:meth:`randomly-determined <attributes.StringAttribute.mutate_value>` member of the aggregation_options. Valid values are in [0.0, 1.0].

.. index:: X_options

• aggregation_options

  A space-separated list of the aggregation functions that may be used by nodes. This defaults to "sum". The available functions (defined in aggregations) are: sum, :py:func:`product <aggregations.product_aggregation>`, min, max, mean, median, and :py:func:`maxabs <aggregations.maxabs_aggregation>` (which returns the input value with the greatest absolute value; the returned value may be positive or negative). New aggregation functions can be defined similarly to :ref:`new activation functions <customization-label>`. (Note that the function needs to take a list or other iterable; the reduce <functools.reduce> function, as in aggregations, may be of use in this.)

  .. versionchanged:: 0.92
    Moved out of :py:mod:`genome` into :py:mod:`aggregations`; maxabs, mean, and median added; method for defining new aggregation functions added.
  
  

.. index:: bias

.. index:: mutation

.. index:: node

.. index:: attributes

.. index:: init_mean

• bias_init_mean

  The mean of the normal/gaussian distribution, if it is used to :py:meth:`select <attributes.FloatAttribute.init_value>` :term:`bias` :term:`attribute <attributes>` values for new :term:`nodes <node>`.

.. index:: init_stdev

• bias_init_stdev

  The standard deviation of the normal/gaussian distribution, if it is used to select bias values for new nodes.

.. index:: init_type

• bias_init_type

  If set to gaussian or normal, then the initialization is to a normal/gaussian distribution. If set to uniform, a uniform distribution from \max(bias\_min\_value, (bias\_init\_mean-(bias\_init\_stdev*2))) to \min(bias\_max\_value, (bias\_init\_mean+(bias\_init\_stdev*2))). (Note that the standard deviation of a uniform distribution is not range/0.25, as implied by this, but the range divided by a bit over 0.288 (the square root of 12); however, this approximation makes setting the range much easier.) This defaults to "gaussian".

  .. versionadded:: 0.92
  
  

.. index:: max_value

.. index:: min_value

• bias_max_value

  The maximum allowed bias value. Biases above this value will be :py:meth:`clamped <attributes.FloatAttribute.clamp>` to this value.

• bias_min_value

  The minimum allowed bias value. Biases below this value will be :py:meth:`clamped <attributes.FloatAttribute.clamp>` to this value.

.. index:: mutate_power

• bias_mutate_power

  The standard deviation of the zero-centered normal/gaussian distribution from which a bias value :term:`mutation` is drawn.

.. index:: mutate_rate

• bias_mutate_rate

  The probability that :term:`mutation` will change the bias of a node by adding a random value.

.. index:: replace_rate

• bias_replace_rate

  The probability that :term:`mutation` will replace the bias of a node with a newly :py:meth:`chosen <attributes.FloatAttribute.init_value>` random value (as if it were a new node).

.. index:: ! compatibility_disjoint_coefficient

.. index:: disjoint

• compatibility_disjoint_coefficient

  The coefficient for the :term:`disjoint` and :term:`excess` :term:`gene` counts' contribution to the :term:`genomic distance`.

.. index:: ! compatibility_weight_coefficient

.. index:: attributes

.. index:: homologous

• compatibility_weight_coefficient

  The coefficient for each :term:`weight`, :term:`bias`, or :term:`response` multiplier difference's contribution to the :term:`genomic distance` (for :term:`homologous` :term:`nodes <node>` or :term:`connections <connection>`). This is also used as the value to add for differences in :term:`activation functions <activation function>`, :term:`aggregation functions <aggregation function>`, or :term:`enabled`/disabled status.

Note

It is currently possible for two :term:`homologous` nodes or connections to have a higher contribution to the :term:`genomic distance` than a disjoint or excess :term:`node` or :term:`connection`, depending on their :term:`attributes` and the settings of the above parameters.

.. index:: mutation

.. index:: connection

.. index:: conn_add_prob

• conn_add_prob

  The probability that :term:`mutation` will add a :term:`connection` between existing :term:`nodes <node>`. Valid values are in [0.0, 1.0].

.. index:: conn_delete_prob

• conn_delete_prob

  The probability that :term:`mutation` will delete an existing connection. Valid values are in [0.0, 1.0].

.. index:: enabled

.. index:: ! enabled_default

.. index:: initial_connection

.. index:: connection

.. index:: attributes

.. index:: X_default

• enabled_default

  The default :term:`enabled` :term:`attribute <attributes>` of newly created connections. Valid values are True and False.

Note

"Newly created connections" include ones in newly-created genomes, if those have initial connections (from the setting of the :ref:`initial_connection <initial-connection-config-label>` variable).

.. index:: mutation

.. index:: mutate_rate

• enabled_mutate_rate

  The probability that :term:`mutation` will :py:func:`replace <attributes.BoolAttribute.mutate_value>` (50/50 chance of True or False) the enabled status of a connection. Valid values are in [0.0, 1.0].

.. index:: rate_to_false_add

.. index:: rate_to_true_add

• enabled_rate_to_false_add

  Adds to the enabled_mutate_rate if the connection is currently :term:`enabled`.

• enabled_rate_to_true_add

  Adds to the enabled_mutate_rate if the connection is currently not enabled.

  .. versionadded:: 0.92
    ``enabled_rate_to_false_add`` and ``enabled_rate_to_true_add``
  
  

.. index:: ! feed_forward

.. index:: feedforward

• feed_forward

  If this evaluates to True, generated networks will not be allowed to have :term:`recurrent` :term:`connections <connection>` (they will be :term:`feedforward`). Otherwise they may be (but are not forced to be) recurrent.

.. index:: ! initial_connection

.. index:: enabled_default

.. index:: connection

• initial_connection

  Specifies the initial connectivity of newly-created genomes. (Note the effects on settings other than unconnected of the :ref:`enabled_default <enabled-default-label>` parameter.) There are seven allowed values:

  • unconnected - No :term:`connections <connection>` are initially present. This is the default.
  • fs_neat_nohidden - One randomly-chosen :term:`input node` has one connection to each :term:`output node`. (This is one version of the FS-NEAT scheme; "FS" stands for "Feature Selection".)
  • fs_neat_hidden - One randomly-chosen :term:`input node` has one connection to each :term:`hidden <hidden node>` and :term:`output node`. (This is another version of the FS-NEAT scheme. If there are no hidden nodes, it is the same as fs_neat_nohidden.)
  • full_nodirect - Each :term:`input node` is connected to all :term:`hidden <hidden node>` nodes, if there are any, and each hidden node is connected to all :term:`output nodes <output node>`; otherwise, each input node is connected to all :term:`output nodes <output node>`. Genomes with :ref:`feed_forward <feed-forward-config-label>` set to False will also have :term:`recurrent` (loopback, in this case) connections from each hidden or output node to itself.
  • full_direct - Each :term:`input node` is connected to all :term:`hidden <hidden node>` and :term:`output nodes <output node>`, and each hidden node is connected to all output nodes. Genomes with :ref:`feed_forward <feed-forward-config-label>` set to False will also have :term:`recurrent` (loopback, in this case) connections from each hidden or output node to itself.
  • partial_nodirect # - As for full_nodirect, but each connection has a probability of being present determined by the number (valid values are in [0.0, 1.0]).
  • partial_direct # - as for full_direct, but each connection has a probability of being present determined by the number (valid values are in [0.0, 1.0]).

.. versionchanged:: 0.92
  fs_neat split into fs_neat_nohidden and fs_neat_hidden; full, partial split into full_nodirect, full_direct, partial_nodirect, partial_direct

.. index:: mutation

.. index:: node

.. index:: node_add_prob

• node_add_prob

  The probability that :term:`mutation` will add a new :term:`node` (essentially replacing an existing connection, the :term:`enabled` status of which will be set to False). Valid values are in [0.0, 1.0].

.. index:: node_delete_prob

• node_delete_prob

  The probability that :term:`mutation` will delete an existing node (and all connections to it). Valid values are in [0.0, 1.0].

.. index:: hidden node

.. index:: ! num_hidden

• num_hidden

  The number of :term:`hidden nodes <hidden node>` to add to each genome in the initial population.

.. index:: input node

.. index:: ! num_inputs

• num_inputs

  The number of :term:`input nodes <input node>`, through which the network receives inputs.

.. index:: output node

.. index:: ! num_outputs

• num_outputs

  The number of :term:`output nodes <output node>`, to which the network delivers outputs.

.. index:: response

.. index:: mutation

.. index:: node

.. index:: attributes

.. index:: init_mean

• response_init_mean

  The mean of the normal/gaussian distribution, if it is used to :py:meth:`select <attributes.FloatAttribute.init_value>` :term:`response` multiplier :term:`attribute <attributes>` values for new :term:`nodes <node>`.

.. index:: init_stdev

• response_init_stdev

  The standard deviation of the normal/gaussian distribution, if it is used to select response multipliers for new nodes.

.. index:: init_type

• response_init_type

  If set to gaussian or normal, then the initialization is to a normal/gaussian distribution. If set to uniform, a uniform distribution from \max(response\_min\_value, (response\_init\_mean-(response\_init\_stdev*2))) to \min(response\_max\_value, (response\_init\_mean+(response\_init\_stdev*2))). (Note that the standard deviation of a uniform distribution is not range/0.25, as implied by this, but the range divided by a bit over 0.288 (the square root of 12); however, this approximation makes setting the range much easier.) This defaults to "gaussian".

  .. versionadded:: 0.92
  
  

.. index:: max_value

.. index:: min_value

• response_max_value

  The maximum allowed response multiplier. Response multipliers above this value will be :py:meth:`clamped <attributes.FloatAttribute.clamp>` to this value.

• response_min_value

  The minimum allowed response multiplier. Response multipliers below this value will be :py:meth:`clamped <attributes.FloatAttribute.clamp>` to this value.

.. index:: mutate_power

• response_mutate_power

  The standard deviation of the zero-centered normal/gaussian distribution from which a response multiplier :term:`mutation` is drawn.

.. index:: mutate_rate

• response_mutate_rate

  The probability that :term:`mutation` will change the response multiplier of a node by adding a random value.

.. index:: replace_rate

• response_replace_rate

  The probability that :term:`mutation` will replace the response multiplier of a node with a newly :py:meth:`chosen <attributes.FloatAttribute.init_value>` random value (as if it were a new node).

.. index:: ! single_structural_mutation

.. index:: ! structural_mutation_surer

.. index:: mutation

.. index:: node

.. index:: connection

• single_structural_mutation

  If this evaluates to True, only one structural mutation (the addition or removal of a :term:`node` or :term:`connection`) will be allowed per genome per generation. (If the probabilities for :ref:`conn_add_prob <conn-add-prob-label>`, conn_delete_prob, :ref:`node_add_prob <node-add-prob-label>`, and node_delete_prob add up to over 1, the chances of each are proportional to the appropriate configuration value.) This defaults to "False".

  .. versionadded:: 0.92
  
  

• structural_mutation_surer

  If this evaluates to True, then an attempt to add a :term:`node` to a genome lacking :term:`connections <connection>` will result in adding a connection instead; furthermore, if an attempt to add a connection tries to add a connection that already exists, that connection will be :term:`enabled`. If this is set to default, then it acts as if it had the same value as single_structural_mutation (above). This defaults to "default".

  .. versionadded:: 0.92
  
  

.. index:: weight

.. index:: mutation

.. index:: connection

.. index:: attributes

.. index:: init_mean

• weight_init_mean

  The mean of the normal/gaussian distribution used to :py:meth:`select <attributes.FloatAttribute.init_value>` :term:`weight` :term:`attribute <attributes>` values for new :term:`connections <connection>`.

.. index:: init_stdev

• weight_init_stdev

  The standard deviation of the normal/gaussian distribution used to select weight values for new connections.

.. index:: ! init_type

• weight_init_type

  If set to gaussian or normal, then the initialization is to a normal/gaussian distribution. If set to uniform, a uniform distribution from \max(weight\_min\_value, (weight\_init\_mean-(weight\_init\_stdev*2))) to \min(weight\_max\_value, (weight\_init\_mean+(weight\_init\_stdev*2))). (Note that the standard deviation of a uniform distribution is not range/0.25, as implied by this, but the range divided by a bit over 0.288 (the square root of 12); however, this approximation makes setting the range much easier.) This defaults to "gaussian".

  .. versionadded:: 0.92
  
  

.. index:: max_value

.. index:: min_value

• weight_max_value

  The maximum allowed weight value. Weights above this value will be :py:meth:`clamped <attributes.FloatAttribute.clamp>` to this value.

• weight_min_value

  The minimum allowed weight value. Weights below this value will be :py:meth:`clamped <attributes.FloatAttribute.clamp>` to this value.

.. index:: mutate_power

• weight_mutate_power

  The standard deviation of the zero-centered normal/gaussian distribution from which a weight value :term:`mutation` is drawn.

.. index:: mutate_rate

• weight_mutate_rate

  The probability that :term:`mutation` will change the weight of a connection by adding a random value.

.. index:: replace_rate

• weight_replace_rate

  The probability that :term:`mutation` will replace the weight of a connection with a newly :py:meth:`chosen <attributes.FloatAttribute.init_value>` random value (as if it were a new connection).

:ref:`Table of Contents <toc-label>`