TransformerMulti

class TransformerMulti(transformers, variable_sizes=None)

Join transformers together for states with multiple variables.

This class is for states that can be written (after discretization) as

\[\begin{split}\q = \left[\begin{array}{c} \q_{0} \\ \q_{1} \\ \vdots \\ \q_{n_q - 1} \end{array}\right] \in \RR^{n},\end{split}\]

where each \(\q_{i} \in \RR^{n_i}\) represents a single discretized state variable of dimension \(n_i \in \NN\). The dimension of the joint state \(\q\) is \(n = \sum_{i=0}^{n_q - 1} n_i\), i.e., state_dimension = sum(variable_sizes). Individual transformers are calibrated for each state variable.

Parameters

transformerslist

Initialized (but not necessarily trained) transformer objects, one for each state variable.

variable_sizeslist or None

Dimensions for each state variable, \(n_0,\ldots,n_{n_q-1}\). If None (default), set \(n_i\) to transformers[i].state_dimension; if any of these are None, assume all state variables have the same dimension, i.e., \(n_0 = n_1 = \cdots = n_x \in \NN\) with \(n_x\) to be determined in fit(). In this case, \(n = n_q n_x\).

Properties

num_variables	Number of state variables \(n_q \in \NN\).
state_dimension	Total dimension \(n = \sum_{i=0}^{n_q-1} n_i \in \NN\) of the joint state.
transformers	Transformers for each state variable.
variable_names	Names for each state variable.
variable_sizes	Dimensions of each state variable, \(n_0,\ldots,n_{n_q-1}\in\NN\).

Methods

fit	Learn (but do not apply) the transformation.
fit_transform	Learn and apply the transformation.
get_var	Extract a single variable from the joint state.
inverse_transform	Apply the inverse of the learned transformation.
load	Load a previously saved transformer from an HDF5 file.
save	Save the transformer to an HDF5 file.
split	Split the joint state into the individual state variables.
transform	Apply the learned transformation.
transform_ddts	Apply the learned transformation to snapshot time derivatives.
verify	Verify that transform() and inverse_transform() are consistent and that transform_ddts(), if implemented in each transformer, is consistent with transform().