SimpleTensor (Stanford CoreNLP API)

java.lang.Object
- edu.stanford.nlp.neural.SimpleTensor

All Implemented Interfaces:

Serializable
```
public class SimpleTensor
extends Object
implements Serializable
```
This class defines a block tensor, somewhat like a three dimensional matrix. This can be created in various ways, such as by providing an array of SimpleMatrix slices, by providing the initial size to create a 0-initialized tensor, or by creating a random matrix.

Author:

John Bauer, Richard Socher

See Also:
Serialized Form

Constructor Summary

Constructors
Constructor and Description
`SimpleTensor(int numRows, int numCols, int numSlices)` Creates a zero initialized tensor
`SimpleTensor(org.ejml.simple.SimpleMatrix[] slices)` Copies the data in the slices.

Method Summary

Methods
Modifier and Type	Method and Description
`org.ejml.simple.SimpleMatrix`	`bilinearProducts(org.ejml.simple.SimpleMatrix in)` Returns a column vector where each entry is the nth bilinear product of the nth slices of the two tensors.
`SimpleTensor`	`elementMult(SimpleTensor other)` Performs elementwise multiplication on the tensors.
`double`	`elementSum()` Returns the sum of all elements in the tensor.
`int`	`getNumElements()` Total number of elements in the tensor
`org.ejml.simple.SimpleMatrix`	`getSlice(int slice)` Returns the SimpleMatrix at `slice`.
`boolean`	`isZero()` Returns true iff every element of the tensor is 0
`Iterator<org.ejml.simple.SimpleMatrix>`	`iteratorSimpleMatrix()` Returns an iterator over the `SimpleMatrix` objects contained in the tensor.
`static Iterator<org.ejml.simple.SimpleMatrix>`	`iteratorSimpleMatrix(Iterator<SimpleTensor> tensors)` Returns an Iterator which returns the SimpleMatrices represented by an Iterator over tensors.
`int`	`numCols()` Number of columns in the tensor
`int`	`numRows()` Number of rows in the tensor
`int`	`numSlices()` Number of slices in the tensor
`SimpleTensor`	`plus(SimpleTensor other)` Returns `other` added to the current object, which is unaffected.
`static SimpleTensor`	`random(int numRows, int numCols, int numSlices, double minValue, double maxValue, Random rand)` Returns a randomly initialized tensor with values draft from the uniform distribution between minValue and maxValue.
`SimpleTensor`	`scale(double scaling)` Returns a new tensor which has the values of the original tensor scaled by `scaling`.
`void`	`set(double value)`
`void`	`setSlice(int slice, org.ejml.simple.SimpleMatrix matrix)` Use the given `matrix` in place of `slice`.
`String`	`toString()`

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

- Constructor Detail
  - SimpleTensor
```
public SimpleTensor(int numRows,
            int numCols,
            int numSlices)
```
    Creates a zero initialized tensor
  - SimpleTensor
```
public SimpleTensor(org.ejml.simple.SimpleMatrix[] slices)
```
    Copies the data in the slices. Slices are copied rather than reusing the original SimpleMatrix objects. Each slice must be the same size.
- Method Detail
  - random
```
public static SimpleTensor random(int numRows,
                  int numCols,
                  int numSlices,
                  double minValue,
                  double maxValue,
                  Random rand)
```
    Returns a randomly initialized tensor with values draft from the uniform distribution between minValue and maxValue.
  - numRows
```
public int numRows()
```
    Number of rows in the tensor
  - numCols
```
public int numCols()
```
    Number of columns in the tensor
  - numSlices
```
public int numSlices()
```
    Number of slices in the tensor
  - getNumElements
```
public int getNumElements()
```
    Total number of elements in the tensor
  - set
```
public void set(double value)
```
  - scale
```
public SimpleTensor scale(double scaling)
```
    Returns a new tensor which has the values of the original tensor scaled by scaling. The original object is unaffected.
  - plus
```
public SimpleTensor plus(SimpleTensor other)
```
    Returns other added to the current object, which is unaffected.
  - elementMult
```
public SimpleTensor elementMult(SimpleTensor other)
```
    Performs elementwise multiplication on the tensors. The original objects are unaffected.
  - elementSum
```
public double elementSum()
```
    Returns the sum of all elements in the tensor.
  - setSlice
```
public void setSlice(int slice,
            org.ejml.simple.SimpleMatrix matrix)
```
    Use the given matrix in place of slice. Does not copy the matrix, but rather uses the actual object.
  - getSlice
```
public org.ejml.simple.SimpleMatrix getSlice(int slice)
```
    Returns the SimpleMatrix at slice.
    The actual slice is returned - do not alter this unless you know what you are doing.
  - bilinearProducts
```
public org.ejml.simple.SimpleMatrix bilinearProducts(org.ejml.simple.SimpleMatrix in)
```
    Returns a column vector where each entry is the nth bilinear product of the nth slices of the two tensors.
  - isZero
```
public boolean isZero()
```
    Returns true iff every element of the tensor is 0
  - iteratorSimpleMatrix
```
public Iterator<org.ejml.simple.SimpleMatrix> iteratorSimpleMatrix()
```
    Returns an iterator over the SimpleMatrix objects contained in the tensor.
  - iteratorSimpleMatrix
```
public static Iterator<org.ejml.simple.SimpleMatrix> iteratorSimpleMatrix(Iterator<SimpleTensor> tensors)
```
    Returns an Iterator which returns the SimpleMatrices represented by an Iterator over tensors. This is useful for if you want to perform some operation on each of the SimpleMatrix slices, such as turning them into a parameter stack.
  - toString
```
public String toString()
```
    Overrides:
    
    toString in class Object

Class SimpleTensor

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Constructor Detail

SimpleTensor

SimpleTensor

Method Detail

random

numRows

numCols

numSlices

getNumElements

set

scale

plus

elementMult

elementSum

setSlice

getSlice

bilinearProducts

isZero

iteratorSimpleMatrix

iteratorSimpleMatrix

toString