FunctionVariable (neureka 1.0.1 API)

java.lang.Object
- neureka.math.implementations.FunctionVariable

All Implemented Interfaces:

Function
```
public final class FunctionVariable
extends java.lang.Object
implements Function
```
Instances of this implementation of the Function interface are leave nodes within the abstract syntax tree of a function, representing indexed inputs to a function. When parsing an expression into a function then these inputs are recognized by the character 'i' or 'I', followed by the character 'j' or 'J' (optionally wrapped by '[' and ']'), which is a placeholder for the index of the argument within the list/array of arguments passed to a concrete Function instance.
So for example, when creating a function by calling the following factory method...

Function.of(java.lang.String)( "3 * sum( (I[j] + 4) * I[0] )" )

...then the substrings "I[j]" will be parsed into instances of this class!
The substring "I[0]" on the other hand will not be parsed into an instance of this class!

Nested Class Summary
- Nested classes/interfaces inherited from interface neureka.math.Function
  Function.Callable

Constructor Summary

Constructors
Constructor and Description

FunctionVariable(java.lang.String equation)

Constructors
Constructor and Description
`FunctionVariable(java.lang.String equation)`

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`double`	`call(double[] inputs, int j)` Invokes this `Function` with the provided array of inputs ad an index for input dependent indexing.
`boolean`	`dependsOn(int index)` Use this to determine if this function directly or indirectly references an input with the provided index.
`double`	`derive(double[] inputs, int index)` Calculates the derivative of a particular input with respect to the output of this `Function` based on the provided array of inputs and an index targeting the input to be derived.
`double`	`derive(double[] inputs, int index, int j)` Calculates the derivative of a particular input with respect to the output of this `Function` based on the provided array of inputs, an index targeting the input to be derived and an index for input dependent indexing.
`Tensor<?>`	`execute(Args arguments, Tensor<?>... inputs)` Warning: Tensors returned by this method are eligible for deletion when consumed by other functions. Use this to call this `Function` alongside with some additional meta-arguments which will be passed to the underlying `Operation`(s).
`Function`	`getDerivative(int index)` This method builds a new `Function` which is the derivative of this `Function` with respect to the provided input index.
`AbstractOperation`	`getOperation()`
`java.util.List<Function>`	`getSubFunctions()`
`boolean`	`isDoingAD()` Only branch `Function`s can do autograd / 'Auto-Differentiation', meaning functions whose `Function.isFlat()` flag is set to false!
`boolean`	`isFlat()`
`boolean`	`providesGradient()`
`java.lang.String`	`toString()` Turns this function into a string representation which can be used to reconstruct this function or combine it with other function strings to parse entirely new functions...

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface neureka.math.Function
call, call, call, call, call, call, call, call, call, derive, derive, derive, derive, execute, execute, execute, executeDerive, executeDerive, getAllFunctions, invoke, invoke, invoke, invoke, invoke, invoke, invoke, invoke, invoke, numberOfArgs, of, of, with, with

- Constructor Detail
  - FunctionVariable
```
public FunctionVariable(java.lang.String equation)
```
- Method Detail
  - providesGradient
```
public boolean providesGradient()
```
    Returns:
    
    The answer of the question : "does this variable/input node return a gradient?"
  - isFlat
```
public boolean isFlat()
```
    Specified by:
    
    isFlat in interface Function
    
    Returns:
    
    The truth value determining if the sub-functions of this Function do not themselves reference Functions.
  - isDoingAD
```
public boolean isDoingAD()
```
    Description copied from interface: Function
    
    Only branch Functions can do autograd / 'Auto-Differentiation', meaning functions whose Function.isFlat() flag is set to false!
    
    Specified by:
    
    isDoingAD in interface Function
    
    Returns:
    
    The truth value determining if this Function can perform autograd/auto-differentiation on the input tensors it receives.
  - getOperation
```
public AbstractOperation getOperation()
```
    Specified by:
    
    getOperation in interface Function
    
    Returns:
    
    The Operation implementation instance responsible for executing any inputs received by this Function or null if this Function.isFlat().
  - dependsOn
```
public boolean dependsOn(int index)
```
    Description copied from interface: Function
    
    Use this to determine if this function directly or indirectly references an input with the provided index.
    
    Specified by:
    
    dependsOn in interface Function
    
    Parameters:
    
    index - The index which ought to match the input index of a potentially referenced FunctionInput.
    
    Returns:
    
    The truth value determining if this Function (or any sub-functions) reference a FunctionInput with the provided index.
  - getDerivative
```
public Function getDerivative(int index)
```
    Description copied from interface: Function
    
    This method builds a new Function which is the derivative of this Function with respect to the provided input index.
    
    Specified by:
    
    getDerivative in interface Function
    
    Parameters:
    
    index - The index of the input which ought to serve as the variable which ought to be derived.
    
    Returns:
    
    The derivative of this Function.
  - getSubFunctions
```
public java.util.List<Function> getSubFunctions()
```
    Specified by:
    
    getSubFunctions in interface Function
    
    Returns:
    
    The referenced child Function nodes of this Function AST node.
  - call
```
public double call(double[] inputs,
                   int j)
```
    Description copied from interface: Function
    
    Invokes this Function with the provided array of inputs ad an index for input dependent indexing. This method is functionally equivalent to Function.invoke(double[], int).
    
    Specified by:
    
    call in interface Function
    
    Parameters:
    
    inputs - The array of inputs.
    
    j - The index for input dependent indexing.
    
    Returns:
    
    The scalar result, a single double value.
  - derive
```
public double derive(double[] inputs,
                     int index)
```
    Description copied from interface: Function
    
    Calculates the derivative of a particular input with respect to the output of this Function based on the provided array of inputs and an index targeting the input to be derived.
    
    Specified by:
    
    derive in interface Function
    
    Parameters:
    
    inputs - The double array of inputs.
    
    index - The index of the input to be derived.
    
    Returns:
    
    The scalar double result, a single double value.
  - derive
```
public double derive(double[] inputs,
                     int index,
                     int j)
```
    Description copied from interface: Function
    
    Calculates the derivative of a particular input with respect to the output of this Function based on the provided array of inputs, an index targeting the input to be derived and an index for input dependent indexing.
    
    Specified by:
    
    derive in interface Function
    
    Parameters:
    
    inputs - The double array of inputs.
    
    index - The index of the input to be derived.
    
    j - The index for input dependent indexing.
    
    Returns:
    
    The scalar double result, a single double value.
  - execute
```
public Tensor<?> execute(Args arguments,
                         Tensor<?>... inputs)
```
    Description copied from interface: Function
    
    Warning: Tensors returned by this method are eligible for deletion when consumed by other functions.
    Use this to call this Function alongside with some additional meta-arguments which will be passed to the underlying Operation(s).
    
    Specified by:
    
    execute in interface Function
    
    Parameters:
    
    arguments - A set of arguments you want to supply to this function for further control over the execution.
    
    inputs - The tensors which should be sent through this function.
    
    Returns:
    
    The resulting tensor produced by this function.
  - toString
```
public java.lang.String toString()
```
    Description copied from interface: Function
    
    Turns this function into a string representation which can be used to reconstruct this function or combine it with other function strings to parse entirely new functions...
    
    Specified by:
    
    toString in interface Function
    
    Overrides:
    
    toString in class java.lang.Object
    
    Returns:
    
    The string representation of this function.

Class FunctionVariable

Nested Class Summary

Nested classes/interfaces inherited from interface neureka.math.Function

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Methods inherited from interface neureka.math.Function

Constructor Detail

FunctionVariable

Method Detail

providesGradient

isFlat

isDoingAD

getOperation

dependsOn

getDerivative

getSubFunctions

call

derive

derive

execute

toString