Package io.deephaven.integrations.learn

Class Computer

java.lang.Object

io.deephaven.integrations.learn.Computer

public class Computer
extends Object

Computer creates a new deferred calculation plus an offset (FutureOffset) every time computer is called.

Constructor Summary

Constructors

Constructor	Description
Computer(Table table, Function<Object[],Object> modelFunc, Input[] inputs, int batchSize)	Creates a new Computer.
Computer(Table table, org.jpy.PyObject modelFunc, Input[] inputs, int batchSize)	Creates a new Computer.

Method Summary

Modifier and Type	Method	Description
boolean	clear()	Resets the current future after each set of calculations.
FutureOffset	compute(long k)	Adds new row indices to be used in the deferred calculation.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

Computer

public Computer(Table table,
 org.jpy.PyObject modelFunc,
 Input[] inputs,
 int batchSize)

Creates a new Computer.

Parameters:

modelFunc - python function to call on the given inputs from a table.

inputs - inputs to the model function.

batchSize - maximum number of rows for each deferred computation.

Computer

public Computer(Table table,
 Function<Object[],Object> modelFunc,
 Input[] inputs,
 int batchSize)

Creates a new Computer.

Parameters:

modelFunc - python function to call on the given inputs from a table.

inputs - inputs to the model function.

batchSize - maximum number of rows for each deferred computation.

Method Details

clear

public boolean clear()

Resets the current future after each set of calculations.

Returns:

always false, because functions used in query strings cannot return nothing.

compute

public FutureOffset compute(long k)

Adds new row indices to be used in the deferred calculation.

Parameters:

k - index to be added to the current index set.

Returns:

future offset that combines a future with the relevant row index to access result.