DeepLearning - Maple Help

Online Help

All Products    Maple    MapleSim


Overview of the DeepLearning Package

  

The DeepLearning package is a collection of tools for machine learning. The package supports several common operations used with neural networks, including classification and regression.

 

DeepLearning Types

Commands for Managing Tensors

Commands for Managing Dataflow Graphs

Commands for Constructing Estimators

Commands for Constructing Feature Columns

Commands for Managing Sessions

Details

Compatibility

DeepLearning Types

• 

DeepLearning makes use of the following custom types

DataflowGraph

Estimator

FeatureColumn

Layer

Model

Model

Model

Model

Model

Optimizer

Session

Tensor

Commands for Managing Tensors

• 

The core object in a DeepLearning computation is a Tensor. The following commands construct Tensor objects in the active graph.

AddMultiple

ApplyOperation

Classify

Concatenate

Constant

EinsteinSummation

Fill

GetEagerExecution

GetVariable

IdentityMatrix

OneHot

Ones

Optimizer

Placeholder

RandomTensor

Sequential

SetEagerExecution

Variable

VariablesInitializer

Zeros

Commands for Managing Dataflow Graphs

• 

The following commands allow querying and modification of the DataflowGraph in which the current computation occurs.

GetDefaultGraph

ResetDefaultGraph

Restore

Save

SetRandomSeed

Variables

 

 

Commands for Constructing Estimators

• 

The following commands construct Estimator objects for classification and regression tasks.

DNNClassifier

DNNLinearCombinedClassifier

DNNLinearCombinedRegressor

DNNRegressor

Estimator

LinearClassifier

LinearRegressor

 

Commands for Constructing Feature Columns

• 

The following commands construct FeatureColumn objects for use with an Estimator.

BucketizedColumn

CategoricalColumn

NumericColumn

 

Commands for Managing Sessions

• 

The following commands manage Session objects.

GetDefaultSession

Session

 

 

Details

• 

The DeepLearning package is implemented using Google TensorFlow™ and provides access to a subset of the TensorFlow Python API, version 2.2.0.

• 

The DeepLearning package is currently not supported on the following platforms: Macs powered by Apple's M1 chip (Apple Silicon).

• 

For Windows, a processor with AVX instructions is required. For more information, see the Release 1.6.0 section in https://github.com/tensorflow/tensorflow/blob/r1.10/RELEASE.md.

Compatibility

• 

The DeepLearning package was introduced in Maple 2018.

• 

For more information on Maple 2018 changes, see Updates in Maple 2018.

See Also

Statistics