gemseo.core.discipline.discipline module#

The discipline class.

class Discipline(name='')[source]#

Bases: BaseDiscipline

The base class for disciplines.

The execute() method is used to do compute output data from input data. The linearize() method can be used to compute the Jacobian of the differentiable outputs with respect to differentiated inputs. The jac stores this Jacobian. This method can evaluate the true derivatives (default) if the _compute_jacobian() method is implemented or approximated the derivatives, depending on the linearization_mode (one of LinearizationMode).

Initialize self. See help(type(self)) for accurate signature.

Parameters:

name (str) --

The name of the discipline. If empty, use the name of the class.

By default it is set to "".

class ApproximationMode(value)#

Bases: StrEnum

The approximation derivation modes.

CENTERED_DIFFERENCES = 'centered_differences'#: The centered differences method used to approximate the Jacobians by perturbing each variable with a small real number.

COMPLEX_STEP = 'complex_step'#: The complex step method used to approximate the Jacobians by perturbing each variable with a small complex number.

FINITE_DIFFERENCES = 'finite_differences'#: The finite differences method used to approximate the Jacobians by perturbing each variable with a small real number.

class InitJacobianType(value)[source]#

Bases: StrEnum

The way to initialize the Jacobian matrices.

DENSE = 'dense'#: Initialized as NumPy arrays filled with zeros.

EMPTY = 'empty'#: Initialized as empty NumPy arrays.

SPARSE = 'sparse'#: Initialized as SciPy CSR arrays filled with zeros.

class LinearizationMode(value)#

Bases: StrEnum

An enumeration.

ADJOINT = 'adjoint'#

AUTO = 'auto'#

CENTERED_DIFFERENCES = 'centered_differences'#

COMPLEX_STEP = 'complex_step'#

DIRECT = 'direct'#

FINITE_DIFFERENCES = 'finite_differences'#

REVERSE = 'reverse'#

add_differentiated_inputs(input_names=())[source]#

Add the inputs against which to differentiate the outputs.

The inputs that do not represent continuous numbers are filtered out.

Parameters:

input_names (Iterable[str]) --

The input variables against which to differentiate the outputs. If empty, use all the inputs.

By default it is set to ().

Raises:

ValueError -- When an input name is not the name of a discipline input.

Return type:

None

add_differentiated_outputs(output_names=())[source]#

Add the outputs to be differentiated.

The outputs that do not represent continuous numbers are filtered out.

Parameters:

output_names (Iterable[str]) --

The outputs to be differentiated. If empty, use all the outputs.

By default it is set to ().

Raises:

ValueError -- When an output name is not the name of a discipline output.

Return type:

None

check_jacobian(input_data=mappingproxy({}), derr_approx=ApproximationMode.FINITE_DIFFERENCES, step=1e-07, threshold=1e-08, linearization_mode=LinearizationMode.AUTO, input_names=(), output_names=(), parallel=False, n_processes=2, use_threading=False, wait_time_between_fork=0, auto_set_step=False, plot_result=False, file_path='jacobian_errors.pdf', show=False, fig_size_x=10, fig_size_y=10, reference_jacobian_path='', save_reference_jacobian=False, indices=mappingproxy({}))[source]#

Check if the analytical Jacobian is correct with respect to a reference one.

If reference_jacobian_path is not None and save_reference_jacobian is True, compute the reference Jacobian with the approximation method and save it in reference_jacobian_path.

If reference_jacobian_path is not None and save_reference_jacobian is False, do not compute the reference Jacobian but read it from reference_jacobian_path.

If reference_jacobian_path is None, compute the reference Jacobian without saving it.

Parameters:

input_data (Mapping[str, ndarray]) --
The input data needed to execute the discipline according to the discipline input grammar. If None, use the Discipline.default_input_data.

By default it is set to {}.
derr_approx (ApproximationMode) --
The approximation method, either "complex_step" or "finite_differences".

By default it is set to "finite_differences".
threshold (float) --
The acceptance threshold for the Jacobian error.

By default it is set to 1e-08.
linearization_mode (LinearizationMode) --
the mode of linearization: direct, adjoint or automated switch depending on dimensions of inputs and outputs (Default value = 'auto')

By default it is set to "auto".
input_names (Iterable[str]) --
The names of the inputs wrt which to differentiate the outputs.

By default it is set to ().
output_names (Iterable[str]) --
The names of the outputs to be differentiated.

By default it is set to ().
step (float) --
The differentiation step.

By default it is set to 1e-07.
parallel (bool) --
Whether to differentiate the discipline in parallel.

By default it is set to False.
n_processes (int) --
The maximum simultaneous number of threads, if use_threading is True, or processes otherwise, used to parallelize the execution.

By default it is set to 2.
use_threading (bool) --
Whether to use threads instead of processes to parallelize the execution; multiprocessing will copy (serialize) all the disciplines, while threading will share all the memory This is important to note if you want to execute the same discipline multiple times, you shall use multiprocessing.

By default it is set to False.
wait_time_between_fork (float) --
The time waited between two forks of the process / thread.

By default it is set to 0.
auto_set_step (bool) --
Whether to compute the optimal step for a forward first order finite differences gradient approximation.

By default it is set to False.
plot_result (bool) --
Whether to plot the result of the validation (computed vs approximated Jacobians).

By default it is set to False.
file_path (str | Path) --
The path to the output file if plot_result is True.

By default it is set to "jacobian_errors.pdf".
show (bool) --
Whether to open the figure.

By default it is set to False.
fig_size_x (float) --
The x-size of the figure in inches.

By default it is set to 10.
fig_size_y (float) --
The y-size of the figure in inches.

By default it is set to 10.
reference_jacobian_path (str | Path) --
The path of the reference Jacobian file.

By default it is set to "".
save_reference_jacobian (bool) --
Whether to save the reference Jacobian.

By default it is set to False.
indices (Mapping[str, int | Sequence[int] | Ellipsis | slice]) --
The indices of the inputs and outputs for the different sub-Jacobian matrices, formatted as {variable_name: variable_components} where variable_components can be either an integer, e.g. 2 a sequence of integers, e.g. [0, 3], a slice, e.g. slice(0,3), the ellipsis symbol (...) or None, which is the same as ellipsis. If a variable name is missing, consider all its components. If None, consider all the components of all the inputs and outputs.

By default it is set to {}.

Returns:

Whether the analytical Jacobian is correct with respect to the reference one.

Return type:

bool

execute(input_data=mappingproxy({}))[source]#

Execute the discipline, i.e. compute output data from input data.

If virtual_execution is True, this method returns the default_output_data. Otherwise, it calls the _run() method performing the true execution and returns the corresponding output data. This _run() method must be implemented in subclasses.

Parameters:

input_data (StrKeyMapping) --

The input data. Complete this dictionary with the default_input_data.

By default it is set to {}.

Returns:

The input and output data.

Return type:

DisciplineData

linearize(input_data=mappingproxy({}), compute_all_jacobians=False, execute=True)[source]#

Compute the Jacobians of some outputs with respect to some inputs.

Parameters:

input_data (StrKeyMapping) --
The input data. If empty, use the `default_input_data.

By default it is set to {}.
compute_all_jacobians (bool) --
Whether to compute the Jacobians of all the outputs with respect to all the inputs. Otherwise, set the output variables to differentiate with add_differentiated_outputs() and the input variables against which to differentiate them with add_differentiated_inputs().

By default it is set to False.
execute (bool) --
Whether to start by executing the discipline to ensure that the discipline was executed with the right input data; it can be almost free if the corresponding output data have been stored in the cache.

By default it is set to True.

Returns:

The Jacobian matrices in the dictionary form {output_name: {input_name: jacobian_matrix}} where jacobian_matrix[i, j] is the partial derivative of output_name[i] wrt input_name[j].

Raises:

ValueError -- When either the inputs for which to differentiate the outputs or the outputs to differentiate are missing.

Return type:

JacobianData

set_jacobian_approximation(jac_approx_type=ApproximationMode.FINITE_DIFFERENCES, jax_approx_step=1e-07, jac_approx_n_processes=1, jac_approx_use_threading=False, jac_approx_wait_time=0)[source]#

Set the Jacobian approximation method.

Sets the linearization mode to approx_method, sets the parameters of the approximation for further use when calling Discipline.linearize().

Parameters:

jac_approx_type (ApproximationMode) --
The approximation method, either "complex_step" or "finite_differences".

By default it is set to "finite_differences".
jax_approx_step (float) --
The differentiation step.

By default it is set to 1e-07.
jac_approx_n_processes (int) --
The maximum simultaneous number of threads, if jac_approx_use_threading is True, or processes otherwise, used to parallelize the execution.

By default it is set to 1.
jac_approx_use_threading (bool) --
Whether to use threads instead of processes to parallelize the execution; multiprocessing will copy (serialize) all the disciplines, while threading will share all the memory This is important to note if you want to execute the same discipline multiple times, you shall use multiprocessing.

By default it is set to False.
jac_approx_wait_time (float) --
The time waited between two forks of the process / thread.

By default it is set to 0.

Return type:

None

set_optimal_fd_step(output_names=(), input_names=(), compute_all_jacobians=False, print_errors=False, numerical_error=2.220446049250313e-16)[source]#

Compute the optimal finite-difference step.

Compute the optimal step for a forward first order finite differences gradient approximation. Requires a first evaluation of the perturbed functions values. The optimal step is reached when the truncation error (cut in the Taylor development), and the numerical cancellation errors (round-off when doing f(x+step)-f(x)) are approximately equal.

Warning

This calls the discipline execution twice per input variables.