DeepLearning[Estimator] - Maple Programming Help

Home : Support : Online Help : Programming : DeepLearning Package : Estimators : Operations with Estimators : DeepLearning/Estimator/Train

DeepLearning[Estimator]

 Train
 train an estimator on specified data

 Calling Sequence est:-Train(dataset,output,opts)

Parameters

 est - an Estimator object dataset - object output - object or none opts - zero or more options as specified below

Options

 • shuffle=truefalse

Indicates whether the training data should be randomly shuffled prior to training.

 • steps=nonnegint or none

The number of training steps to take. Each step selects a sample from dataset to train.

Description

 • The Train(dataset,output,opts) command trains an Estimator against training data specified in dataset.
 • For examples, see Estimator.
 • This operation creates a new Tensor by replicating input

Compatibility

 • The DeepLearning[Estimator][Train] command was introduced in Maple 2018.