An Estimator is an object which encapsulates a high-level interface which encapsulates tasks for training, evaluation, and prediction with machine learning models.

To construct an Estimator object encapsulating a certain classification or regression task, see the DeepLearning Overview section on Estimators.

The following functions can be performed with an Estimator.

$N≔1000\:$

$X≔\mathrm{Statistics}:-\mathrm{RandomVariable}\left('\mathrm{Uniform}\left(-1.\,1.\right)'\right)\:$

$\mathrm{training\_data}≔\mathrm{DataFrame}\left(\mathrm{Statistics}:-\mathrm{Sample}\left(X\,\left[N\,2\right]\right)\,\mathrm{columns}=\left[''x''\,''y''\right]\right)\:$

$\mathrm{class}≔\mathrm{DataSeries}\left(⟨\mathrm{seq}\left(\mathrm{`if`}\left({\mathrm{training\_data}\left[''x''\right]\left[i\right]}^2+{\mathrm{training\_data}\left[''y''\right]\left[i\right]}^2<1\&comma;1\&comma;0\right)\&comma;i=1..N\right)⟩\right)\&colon;$

$\mathrm{with}\left(\mathrm{DeepLearning}\right)\&colon;$

$\mathrm{fc}≔\left[\mathrm{seq}\left(\mathrm{NumericColumn}\left(u\&comma;\mathrm{shape}=\left[1\right]\right)\&comma;u\mathbf{in}\left[''x''\&comma;''y''\right]\right)\right]$

$\mathrm{fc}≔\left[\left[\begin{array}{c}\mathrm{Feature\; Column}\\ \mathrm{NumericColumn(key=\text{'}x\text{'},\; shape=(1,),\; default\_value=None,\; dtype=tf.float32,\; normalizer\_fn=None)}\end{array}\right]\&comma;\left[\begin{array}{c}\mathrm{Feature\; Column}\\ \mathrm{NumericColumn(key=\text{'}y\text{'},\; shape=(1,),\; default\_value=None,\; dtype=tf.float32,\; normalizer\_fn=None)}\end{array}\right]\right]$

$\mathrm{classifier}≔\mathrm{DNNClassifier}\left(\mathrm{fc}\&comma;\mathrm{hidden\_units}=\left[10\&comma;20\&comma;10\right]\&comma;\mathrm{num\_classes}=2\right)$

$\mathrm{classifier}≔\left[\begin{array}{c}\mathrm{DeepLearning\; Estimator}\\ \mathrm{<tensorflow\_estimator.python.estimator.canned.dnn.DNNClassifierV2\; object\; at\; 0x7faf5eeb84f0>}\end{array}\right]$

$\mathrm{classifier}:-\mathrm{Train}\left(\mathrm{training\_data}\&comma;\mathrm{class}\&comma;\mathrm{steps}=2000\&comma;\mathrm{num\_epochs}=\mathrm{none}\&comma;\mathrm{shuffle}=\mathrm{true}\right)$

$\left[\begin{array}{c}\mathrm{DeepLearning\; Tensor}\\ \mathrm{Name:\; none}\\ \mathrm{Shape:\; undefined}\\ \mathrm{Data\; Type:\; undefined}\end{array}\right]$

$\mathrm{test\_data}≔\mathrm{DataFrame}\left(\mathrm{Statistics}:-\mathrm{Sample}\left(X\&comma;\left[5\&comma;2\right]\right)\&comma;\mathrm{columns}=\left[''x''\&comma;''y''\right]\right)\&colon;$

$\mathrm{result}≔\mathrm{classifier}:-\mathrm{Predict}\left(\mathrm{test\_data}\&comma;\mathrm{num\_epochs}=1\&comma;\mathrm{shuffle}=\mathrm{false}\right)\&colon;$

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

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