AEPMM

parametricmatrixmodels.modules.AEPMM

AEPMM

add	Append a module to the end of the model.
add_module	Append a module to the end of the model.
append	Append a module to the end of the model.
append_module	Append a module to the end of the model.
astype	Convenience wrapper to set_precision using the dtype argument, returns self.
compile	Compile the model for training by compiling each module.
copy	Create a deep copy of the module.
deepcopy	Create a deep copy of the module.
deserialize	Deserialize the model from a dictionary.
freeze	Freeze the module parameters by setting trainable to False.
from_file	Load a model from a file and return an instance of the Model class.
get_hyperparameters	Get the hyperparameters of the model as a dictionary.
get_modules	Get the modules of the model.
get_num_trainable_floats	Returns the number of trainable floats in the module.
get_output_shape	Get the output shape of the model given an input shape.
get_params	Get the parameters of the model.
get_rng
get_state	Get the state of the model.
get_trainable_modules	Return a PyTree with the same structure as the model's modules, where each leaf node is a boolean indicating whether the corresponding module is trainable (i.e. mod.trainable == True and mod.get_params() is not None/()).
grad_input	Doc TODO
grad_params	Doc TODO
insert	Insert a module at a specific index in the model.
insert_module	Insert a module at a specific index in the model.
is_ready	Return True if the module is initialized and ready for training or inference.
load	Load the model from a file.
pop	Remove and return a module by key or index in the model.
pop_module_by_index	Remove and return a module at a specific index in the model.
pop_module_by_key	Remove and return a module by key or index in the model.
predict	Call the model with the input data.
prepend	Prepend a module to the beginning of the model.
prepend_module	Prepend a module to the beginning of the model.
reset
save	Save the model to a file.
save_compressed	Save the model to a compressed file.
serialize	Serialize the model to a dictionary.
set_hyperparameters	Set the hyperparameters of the model from a dictionary.
set_params	Set the parameters of the model from a PyTree of PyTrees of numpy arrays.
set_precision	Set the precision of the model parameters and states.
set_rng	Set the random key for the model.
set_state	Set the state of the model from a PyTree of PyTrees of numpy arrays.
train
unfreeze	Unfreeze the module parameters by setting trainable to True.
upgrade	Upgrade serialized module data to the current version.

name	Returns the name of the module, unless overridden, this is the class name.
trainable	Whether the module is trainable (i.e., whether its parameters should be updated during training).

alias of AffineEigenvaluePMM