Methods

Summary

method

class parent

truncated documentation

__getstate__

BaseEstimator

Removes any non pickable attribute.

__getstate__

BaseLearningOnnx

Overwrites getstate to get rid of InferenceSession.

__getstate__

OrtDataLoader

Removes any non pickable attribute.

__getstate__

OrtGradientForwardBackwardOptimizer

Removes any non pickable attribute.

__getstate__

OrtGradientForwardBackward

Removes any non pickable attribute.

__init__

BaseEstimator

__init__

BaseLearningOnnx

__init__

OrtDataLoader

__init__

OrtGradientOptimizer

__init__

OrtGradientForwardBackwardOptimizer

__init__

OrtGradientForwardBackward

__init__

OrtGradientForwardBackwardFunction

__init__

AbsoluteLearningLoss

__init__

BaseLearningLoss

__init__

ElasticLearningLoss

__init__

NegLogLearningLoss

__init__

SquareLearningLoss

__init__

BaseLearningPenalty

__init__

ElasticLearningPenalty

__init__

NoLearningPenalty

__init__

BaseLearningRate

__init__

LearningRateSGD

__init__

LearningRateSGDNesterov

__len__

OrtDataLoader

Returns the number of observations.

__repr__

BaseEstimator

Usual.

__repr__

BaseLearningOnnx

Usual

__repr__

OrtDataLoader

usual

__repr__

OrtGradientForwardBackward

usual

__repr_extended__

BaseLearningOnnx

__repr_extended__

BaseLearningRate

__repr_extended__

LearningRateSGD

__repr_extended__

LearningRateSGDNesterov

__setstate__

BaseEstimator

Restores any non pickable attribute.

__setstate__

BaseLearningOnnx

Overwrites getstate to get rid of InferenceSession.

__setstate__

OrtDataLoader

Restores any non pickable attribute.

__setstate__

OrtGradientForwardBackwardOptimizer

Restores any non pickable attribute.

__setstate__

OrtGradientForwardBackward

Restores any non pickable attribute.

_bind_input_ortvalue

BaseLearningOnnx

Binds C_OrtValue to the structure used by InferenceSession to run inference.

_bind_input_ortvalue

OrtGradientOptimizer

Binds C_OrtValue to the structure used by InferenceSession to run inference.

_bind_output_ortvalue

BaseLearningOnnx

Binds C_OrtValue to the structure used by InferenceSession to run inference.

_call_iobinding

AbsoluteLearningLoss

_call_iobinding

BaseLearningLoss

_call_iobinding

ElasticLearningLoss

_call_iobinding

NegLogLearningLoss

_call_iobinding

SquareLearningLoss

_call_iobinding

BaseLearningPenalty

_call_iobinding

ElasticLearningPenalty

_call_iobinding

NoLearningPenalty

_call_iobinding

BaseLearningRate

_call_iobinding

LearningRateSGD

_call_iobinding

LearningRateSGDNesterov

_create_onnx_graphs

OrtGradientForwardBackward

Creates forward and backward ONNX graph. The new class has the following attributes:

_create_training_session

OrtGradientOptimizer

Creates an instance of TrainingSession.

_create_training_session

OrtGradientForwardBackwardOptimizer

_evaluation

OrtGradientOptimizer

_evaluation

OrtGradientForwardBackwardOptimizer

_get_att_state

OrtGradientForwardBackwardOptimizer

_get_trained_onnx

BaseEstimator

Returns the trained onnx graph, the initial graph modified by replacing the initializers with the trained …

_init_next

OrtGradientForwardBackward

_iteration

OrtGradientOptimizer

_iteration

OrtGradientForwardBackwardOptimizer

_next_iter

OrtDataLoader

backward

OrtGradientForwardBackwardFunction

Implements backward function. The function returns an OrtValueVector.

build_onnx_function

BaseLearningOnnx

This class updates the weights. It assumes it can do operator on OrtValue. This can be done through ONNX …

build_onnx_function

OrtGradientForwardBackwardOptimizer

Creates ONNX graph and InferenceSession related to any operations applying on OrtValue.

build_onnx_function

AbsoluteLearningLoss

build_onnx_function

ElasticLearningLoss

build_onnx_function

NegLogLearningLoss

build_onnx_function

SquareLearningLoss

build_onnx_function

ElasticLearningPenalty

build_onnx_function

NoLearningPenalty

build_onnx_function

LearningRateSGD

build_onnx_function

LearningRateSGDNesterov

build_onnx_score_function

AbsoluteLearningLoss

Assuming the loss function was created. This one takes the onnx graph and generate the onnx graph for the …

build_onnx_score_function

BaseLearningLoss

Assuming the loss function was created. This one takes the onnx graph and generate the onnx graph for the …

build_onnx_score_function

ElasticLearningLoss

Assuming the loss function was created. This one takes the onnx graph and generate the onnx graph for the …

build_onnx_score_function

NegLogLearningLoss

Assuming the loss function was created. This one takes the onnx graph and generate the onnx graph for the …

build_onnx_score_function

SquareLearningLoss

Assuming the loss function was created. This one takes the onnx graph and generate the onnx graph for the …

clear_binding_inputs

BaseLearningOnnx

Clears binding and empty cache.

fit

OrtGradientOptimizer

Trains the model.

fit

OrtGradientForwardBackwardOptimizer

Trains the model.

forward

OrtGradientForwardBackwardFunction

Implements forward function.

get_full_state

OrtGradientForwardBackwardOptimizer

Returns the trained weights and the inputs.

get_initializer

OrtGradientForwardBackward

Returns an initializer as numpy arrays.

get_params

BaseEstimator

Returns the list of parameters. Parameter deep is unused.

get_state

OrtGradientOptimizer

Returns the trained weights.

get_state

OrtGradientForwardBackwardOptimizer

Returns the trained weights.

get_trained_onnx

BaseEstimator

Returns the trained onnx graph, the initial graph modified by replacing the initializers with the trained …

get_trained_onnx

OrtGradientOptimizer

Returns the trained onnx graph, the initial graph modified by replacing the initializers with the trained …

get_trained_onnx

OrtGradientForwardBackwardOptimizer

Returns the trained onnx graph, the initial graph modified by replacing the initializers with the trained …

init_learning_rate

BaseLearningRate

Initializes the learning rate at the beginning of the training.

init_learning_rate

LearningRateSGD

Updates the learning rate at the end of an iteration.

init_learning_rate

LearningRateSGDNesterov

Updates the learning rate at the end of an iteration.

iter_bind

OrtDataLoader

Iterates over the datasets by drawing batch_size consecutive observations. Modifies a bind structure.

iter_numpy

OrtDataLoader

Iterates over the datasets by drawing batch_size consecutive observations. This iterator is slow as it …

iter_ortvalue

OrtDataLoader

Iterates over the datasets by drawing batch_size consecutive observations. This iterator is slow as it …

loop

BaseLearningRate

Loops over learning rate values, n to be precise.

loop

LearningRateSGD

Loops over learning rate values, n to be precise.

loop

LearningRateSGDNesterov

Loops over learning rate values, n to be precise.

loss_gradient

AbsoluteLearningLoss

Returns the loss and the gradient as OrtValue.

loss_gradient

BaseLearningLoss

Returns the loss and the gradient as OrtValue.

loss_gradient

ElasticLearningLoss

Returns the loss and the gradient as OrtValue.

loss_gradient

NegLogLearningLoss

Returns the loss and the gradient as OrtValue.

loss_gradient

SquareLearningLoss

Returns the loss and the gradient as OrtValue.

loss_scores

AbsoluteLearningLoss

Returns the weighted loss (or score) for every observation as OrtValue.

loss_scores

BaseLearningLoss

Returns the weighted loss (or score) for every observation as OrtValue.

loss_scores

ElasticLearningLoss

Returns the weighted loss (or score) for every observation as OrtValue.

loss_scores

NegLogLearningLoss

Returns the weighted loss (or score) for every observation as OrtValue.

loss_scores

SquareLearningLoss

Returns the weighted loss (or score) for every observation as OrtValue.

losses

OrtGradientForwardBackwardOptimizer

Returns the losses associated to every observation.

new_instance

OrtGradientForwardBackward

Creates an instance of class self.cls_type_. It implements methods forward and backward.

penalty_loss

BaseLearningPenalty

Returns the received loss. Updates the loss inplace.

penalty_loss

ElasticLearningPenalty

Computes the penalty associated to every weights and adds them up to the loss.

penalty_loss

NoLearningPenalty

Returns the received loss. Updates the loss inplace.

save_for_backward

OrtGradientForwardBackwardFunction

Saves inputs furing forward steps. The list inputs is copied (simple copy, no deep copy).

save_onnx_graph

BaseOnnxClass

Saves all ONNX files stored in this class.

score

OrtGradientForwardBackwardOptimizer

Return the whole score associated.

set_params

BaseEstimator

Returns the list of parameters. Parameter deep is unused.

set_state

OrtGradientOptimizer

Changes the trained weights.

set_state

OrtGradientForwardBackwardOptimizer

Changes the trained weights.

update_learning_rate

BaseLearningRate

Updates the learning rate at the end of an iteration.

update_learning_rate

LearningRateSGD

Updates the learning rate at the end of an iteration.

update_learning_rate

LearningRateSGDNesterov

Updates the learning rate at the end of an iteration.

update_weights

BaseLearningPenalty

Returns the received loss. Updates the weight inplace.

update_weights

ElasticLearningPenalty

update_weights

NoLearningPenalty

Returns the received loss. Updates the weight inplace.

update_weights

BaseLearningRate

Updates weights based on the algorithm this class is setting up.

update_weights

LearningRateSGD

update_weights

LearningRateSGDNesterov