Class/Object

org.apache.mxnet

NDArray

Related Docs: object NDArray | package mxnet

Permalink

class NDArray extends NativeResource

NDArray object in mxnet. NDArray is basic ndarray/Tensor like data structure in mxnet.
NOTE: NDArray is stored in native memory. Use NDArray in a try-with-resources() construct or a [[org.apache.mxnet.ResourceScope]] in a try-with-resource to have them automatically disposed. You can explicitly control the lifetime of NDArray by calling dispose manually. Failure to do this will result in leaking native memory.

Linear Supertypes
NativeResource, WarnIfNotDisposed, AutoCloseable, AnyRef, Any
Known Subclasses
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. NDArray
  2. NativeResource
  3. WarnIfNotDisposed
  4. AutoCloseable
  5. AnyRef
  6. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new NDArray(handle: NDArrayHandle, writable: Boolean = true, addToCollector: Boolean = true)

    Permalink
    Annotations
    @deprecated
    Deprecated

    (Since version 1.5.0) Please use ResourceScope instead

Value Members

  1. final def !=(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  2. final def ##(): Int

    Permalink
    Definition Classes
    AnyRef → Any
  3. def %(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  4. def %(other: NDArray): NDArray

    Permalink
  5. def %=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  6. def %=(other: NDArray): NDArray

    Permalink
  7. def *(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  8. def *(other: NDArray): NDArray

    Permalink
  9. def **(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  10. def **(other: NDArray): NDArray

    Permalink
  11. def **=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  12. def **=(other: NDArray): NDArray

    Permalink
  13. def *=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  14. def *=(other: NDArray): NDArray

    Permalink
  15. def +(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  16. def +(other: NDArray): NDArray

    Permalink
  17. def +=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  18. def +=(other: NDArray): NDArray

    Permalink
  19. def -(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  20. def -(other: NDArray): NDArray

    Permalink
  21. def -=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  22. def -=(other: NDArray): NDArray

    Permalink
  23. def /(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  24. def /(other: NDArray): NDArray

    Permalink
  25. def /=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  26. def /=(other: NDArray): NDArray

    Permalink
  27. def <(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  28. def <(other: NDArray): NDArray

    Permalink
  29. def <=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  30. def <=(other: NDArray): NDArray

    Permalink
  31. final def ==(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  32. def >(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  33. def >(other: NDArray): NDArray

    Permalink
  34. def >=(other: MX_PRIMITIVE_TYPE): NDArray

    Permalink
  35. def >=(other: NDArray): NDArray

    Permalink
  36. def T: NDArray

    Permalink
  37. def asInContext(context: Context): NDArray

    Permalink

    Return an NDArray that lives in the target context.

    Return an NDArray that lives in the target context. If the array is already in that context, self is returned. Otherwise, a copy is made.

    context

    The target context we want the return value to live in.

    returns

    A copy or self as an NDArray that lives in the target context.

  38. final def asInstanceOf[T0]: T0

    Permalink
    Definition Classes
    Any
  39. def asType(dtype: DType.DType): NDArray

    Permalink

    Return a copied numpy array of current array with specified type.

    Return a copied numpy array of current array with specified type.

    dtype

    Desired type of result array.

    returns

    A copy of array content.

  40. def at(idx: Int): NDArray

    Permalink

    Return a sub NDArray that shares memory with current one.

    Return a sub NDArray that shares memory with current one. the first axis will be rolled up, which causes its shape different from slice(i, i+1)

    idx

    index of sub array.

  41. val bytesAllocated: Long

    Permalink

    Off-Heap Bytes Allocated for this object

    Off-Heap Bytes Allocated for this object

    Definition Classes
    NDArray → NativeResource
  42. def clone(): AnyRef

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  43. def close(): Unit

    Permalink
    Definition Classes
    NativeResource → AutoCloseable
  44. def context: Context

    Permalink

    Get context of current NDArray.

    Get context of current NDArray.

    returns

    The context of current NDArray.

  45. def copy(): NDArray

    Permalink

    Clone the current array

    Clone the current array

    returns

    the copied NDArray in the same context

  46. def copyTo(ctx: Context): NDArray

    Permalink

    Copy the content of current array to a new NDArray in the context.

    Copy the content of current array to a new NDArray in the context.

    ctx

    Target context we want to copy data to.

    returns

    The copy target NDArray

  47. def copyTo(other: NDArray): NDArray

    Permalink

    Copy the content of current array to other.

    Copy the content of current array to other.

    other

    Target NDArray or context we want to copy data to.

    returns

    The copy target NDArray

  48. val creationTrace: Option[Array[StackTraceElement]]

    Permalink
    Attributes
    protected
    Definition Classes
    WarnIfNotDisposed
  49. def dispose(): Unit

    Permalink

    Release the native memory.

    Release the native memory.
    The NDArrays it depends on will NOT be disposed.
    The object shall never be used after it is disposed.

    Definition Classes
    NDArray → NativeResource
  50. def disposeDeps(): NDArray

    Permalink

    Dispose all NDArrays who help to construct this array.

    Dispose all NDArrays who help to construct this array.
    e.g. (a * b + c).disposeDeps() will dispose a, b, c (including their deps) and a * b

    returns

    this NDArray

  51. def disposeDepsExcept(arrs: NDArray*): NDArray

    Permalink

    Dispose all NDArrays who help to construct this array, excepts those in the arguments.

    Dispose all NDArrays who help to construct this array, excepts those in the arguments.
    e.g. (a * b + c).disposeDepsExcept(a, b) will dispose c and a * b. Note that a, b's dependencies will not be disposed either.

    arrs

    array of NDArrays

    returns

    this array

  52. def dtype: DType.DType

    Permalink

    Get data type of current NDArray.

    Get data type of current NDArray.

    returns

    class representing type of current ndarray

  53. final def eq(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  54. def equals(o: Any): Boolean

    Permalink
    Definition Classes
    NDArray → AnyRef → Any
  55. def finalize(): Unit

    Permalink
    Attributes
    protected
    Definition Classes
    WarnIfNotDisposed → AnyRef
  56. final def getClass(): Class[_]

    Permalink
    Definition Classes
    AnyRef → Any
  57. def hashCode(): Int

    Permalink
    Definition Classes
    NDArray → AnyRef → Any
  58. def internal: NDArrayInternal

    Permalink
  59. def isDisposed: Boolean

    Permalink
    Definition Classes
    NativeResource → WarnIfNotDisposed
  60. final def isInstanceOf[T0]: Boolean

    Permalink
    Definition Classes
    Any
  61. def isSparse: Boolean

    Permalink

    check if NDArray is SparseNDArray

    check if NDArray is SparseNDArray

    returns

    Boolean

  62. def logDisposeWarning(): Unit

    Permalink
    Attributes
    protected
    Definition Classes
    WarnIfNotDisposed
  63. def nativeAddress: CPtrAddress

    Permalink

    native Address associated with this object

    native Address associated with this object

    Definition Classes
    NDArray → NativeResource
  64. def nativeDeAllocator: (CPtrAddress) ⇒ Int

    Permalink

    Function Pointer to the NativeDeAllocator of nativeAddress

    Function Pointer to the NativeDeAllocator of nativeAddress

    Definition Classes
    NDArray → NativeResource
  65. final def ne(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  66. final def notify(): Unit

    Permalink
    Definition Classes
    AnyRef
  67. final def notifyAll(): Unit

    Permalink
    Definition Classes
    AnyRef
  68. val ref: NativeResourceRef

    Permalink

    Call NativeResource.register to get the reference

    Call NativeResource.register to get the reference

    Definition Classes
    NDArray → NativeResource
  69. def register(): NativeResourceRef

    Permalink

    Register this object for PhantomReference tracking and in ResourceScope if used inside ResourceScope.

    Register this object for PhantomReference tracking and in ResourceScope if used inside ResourceScope.

    returns

    NativeResourceRef that tracks reachability of this object using PhantomReference

    Definition Classes
    NativeResource
  70. def reshape(dims: Shape): NDArray

    Permalink

    Return a reshaped NDArray that shares memory with current one.

    Return a reshaped NDArray that shares memory with current one.

    dims

    New shape.

    returns

    a reshaped NDArray that shares memory with current one.

  71. def reshape(dims: Array[Long], reverse: Option[Boolean] = None): NDArray

    Permalink

    Return a reshaped NDArray that shares memory with current one.

    Return a reshaped NDArray that shares memory with current one.

    dims

    New shape.

    reverse

    whether to inplace reshape

    returns

    a reshaped NDArray that shares memory with current one.

  72. def reshape(dims: Array[Int]): NDArray

    Permalink

    Return a reshaped NDArray that shares memory with current one.

    Return a reshaped NDArray that shares memory with current one.

    dims

    New shape.

    returns

    a reshaped NDArray that shares memory with current one.

  73. def serialize(): Array[Byte]

    Permalink
  74. def set(other: Array[Double]): NDArray

    Permalink
  75. def set(other: Array[Float]): NDArray

    Permalink
  76. def set(other: NDArray): NDArray

    Permalink
  77. def set(value: MX_PRIMITIVE_TYPE): NDArray

    Permalink

    Set the values of the NDArray

    Set the values of the NDArray

    value

    Value to set

    returns

    Current NDArray

  78. def shape: Shape

    Permalink

    Get shape of current NDArray.

    Get shape of current NDArray.

    returns

    an array representing shape of current ndarray

  79. def size: Int

    Permalink
  80. def slice(i: Int): NDArray

    Permalink

    Return a sliced NDArray at the ith position of axis0

    Return a sliced NDArray at the ith position of axis0

    returns

    a sliced NDArray that shares memory with current one.

  81. def slice(range: (Int, Int)): NDArray

    Permalink
  82. def slice(start: Int, stop: Int): NDArray

    Permalink

    Return a sliced NDArray that shares memory with current one.

    Return a sliced NDArray that shares memory with current one. NDArray only support continuous slicing on axis 0

    start

    Starting index of slice.

    stop

    Finishing index of slice.

    returns

    a sliced NDArray that shares memory with current one.

  83. lazy val sparseFormat: SparseFormat.SparseFormat

    Permalink
  84. final def synchronized[T0](arg0: ⇒ T0): T0

    Permalink
    Definition Classes
    AnyRef
  85. def toArray: Array[Float]

    Permalink

    Return a copied flat java array of current array (row-major).

    Return a copied flat java array of current array (row-major).

    returns

    A copy of array content.

  86. def toFloat64Array: Array[Double]

    Permalink

    Return a copied flat java array of current array (row-major) with datatype as Float64/Double.

    Return a copied flat java array of current array (row-major) with datatype as Float64/Double.

    returns

    A copy of array content.

  87. def toFloat64Scalar: Double

    Permalink
  88. def toScalar: Float

    Permalink

    Return a CPU scalar(float) of current ndarray.

    Return a CPU scalar(float) of current ndarray. This ndarray must have shape (1,)

    returns

    The scalar representation of the ndarray.

  89. def toSparse(sfOption: Option[SparseFormat.SparseFormat] = None): SparseNDArray

    Permalink

    Convert a NDArray to SparseNDArray

    Convert a NDArray to SparseNDArray

    sfOption

    the target sparse type

    returns

    SparseNDArray

  90. def toString(): String

    Permalink

    Visualize the internal structure of NDArray

    Visualize the internal structure of NDArray

    returns

    String that show the structure

    Definition Classes
    NDArray → AnyRef → Any
  91. def tracingEnabled: Boolean

    Permalink
    Attributes
    protected
    Definition Classes
    WarnIfNotDisposed
  92. def unary_-(): NDArray

    Permalink
  93. final def wait(): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  94. final def wait(arg0: Long, arg1: Int): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  95. final def wait(arg0: Long): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  96. def waitToRead(): Unit

    Permalink

    Block until all pending writes operations on current NDArray are finished.

    Block until all pending writes operations on current NDArray are finished. This function will return when all the pending writes to the current NDArray finishes. There can still be pending read going on when the function returns.

  97. val writable: Boolean

    Permalink

Inherited from NativeResource

Inherited from WarnIfNotDisposed

Inherited from AutoCloseable

Inherited from AnyRef

Inherited from Any

Ungrouped