On this page

runtime.integrator settings

Settings namespaces Namespace

Namespace hierarchy

Settings

bin_number_evolution

#

runtime.integrator.bin_number_evolution

Optional per-iteration bin counts, overriding the fixed n_bins value as training proceeds.

Type
array | null
Required
no
Default
Not set
Allowed values
Any value accepted by the declared type

continuous_dim_learning_rate

#

runtime.integrator.continuous_dim_learning_rate

Update rate for adaptive grids over continuous integration dimensions.

Type
number
Required
no
Default
1.0
Allowed values
Any value accepted by the declared type

discrete_dim_learning_rate

#

runtime.integrator.discrete_dim_learning_rate

Update rate for learned probabilities over discrete sampling choices.

Type
number
Required
no
Default
1.0
Allowed values
Any value accepted by the declared type

integrated_phase

#

runtime.integrator.integrated_phase

Complex component to integrate: real, imaginary, or both.

Type
IntegratedPhase
Required
no
Default
"real"
Allowed values
both, imag, real

max_prob_ratio

#

runtime.integrator.max_prob_ratio

Cap the ratio used when turning sampled weights into adaptive probabilities.

Type
number
Required
no
Default
30.0
Allowed values
Any value accepted by the declared type

min_samples_for_update

#

runtime.integrator.min_samples_for_update

Minimum accumulated samples before an adaptive grid or discrete probability is updated.

Type
integer
Required
no
Default
1000
Allowed values
Any value accepted by the declared type

n_bins

#

runtime.integrator.n_bins

Number of bins maintained along each continuous integration dimension.

Type
integer
Required
no
Default
64
Allowed values
Any value accepted by the declared type

n_increase

#

runtime.integrator.n_increase

Samples added to each successive Monte Carlo iteration.

Type
integer
Required
no
Default
10000
Allowed values
Any value accepted by the declared type

n_max

#

runtime.integrator.n_max

Hard upper bound on the total number of integrand evaluations.

Type
integer
Required
no
Default
10000000000
Allowed values
Any value accepted by the declared type

n_start

#

runtime.integrator.n_start

Number of samples in the first Monte Carlo iteration.

Type
integer
Required
no
Default
100000
Allowed values
Any value accepted by the declared type

seed

#

runtime.integrator.seed

Seed for deterministic initialization of Monte Carlo random-number streams.

Type
integer
Required
no
Default
69
Allowed values
Any value accepted by the declared type

show_max_wgt_info

#

runtime.integrator.show_max_wgt_info

Include maximum-weight diagnostics in integration progress and result data.

Type
boolean
Required
no
Default
true
Allowed values
Any value accepted by the declared type

target_absolute_accuracy

#

runtime.integrator.target_absolute_accuracy

Stop once the estimated absolute uncertainty reaches this value, if configured.

Type
number | null
Required
no
Default
Not set
Allowed values
Any value accepted by the declared type

target_relative_accuracy

#

runtime.integrator.target_relative_accuracy

Stop once the estimated relative uncertainty reaches this value, if configured.

Type
number | null
Required
no
Default
Not set
Allowed values
Any value accepted by the declared type

train_on_avg

#

runtime.integrator.train_on_avg

Train adaptive distributions on iteration averages instead of individual sample weights.

Type
boolean
Required
no
Default
false
Allowed values
Any value accepted by the declared type