org.apache

mxnet

package mxnet

Visibility
  1. Public
  2. All

Type Members

  1. class Accuracy extends EvalMetric

    Calculate accuracy

  2. trait BatchEndCallback extends AnyRef

  3. class CompositeEvalMetric extends EvalMetric

    Manage multiple evaluation metrics.

  4. class Context extends Serializable

    Constructing a context which is used to specify the device and device type that will be utilized by the engine.

  5. class CustomMetric extends EvalMetric

    Custom evaluation metric that takes a NDArray function.

  6. abstract class CustomOp extends AnyRef

    Base class for operators implemented in Scala

  7. abstract class CustomOpProp extends AnyRef

    Base class for operator property class implemented in Scala.

    Base class for operator property class implemented in Scala. MXNET_CPU_WORKER_NTHREADS must be greater than 1 for custom op to work on CPU

  8. class DataBatch extends AnyRef

    class batch of data

  9. case class DataDesc(name: String, shape: Shape, dtype: DType.DType = DType.Float32, layout: String = Layout.UNDEFINED) extends Product with Serializable

  10. abstract class DataIter extends Iterator[DataBatch]

    DataIter object in mxnet.

  11. abstract class DataPack extends Iterable[DataBatch]

    pack of DataIter, use as Iterable class

  12. trait EpochEndCallback extends AnyRef

  13. abstract class EvalMetric extends AnyRef

    Base class of all evaluation metrics

  14. class Executor extends NativeResource

    Symbolic Executor component of MXNet
    WARNING: it is your responsibility to clear this object through dispose().

Symbolic Executor component of MXNet
WARNING: it is your responsibility to clear this object through dispose().

See also

Symbol.bind : to create executor

  • class F1 extends EvalMetric

    Calculate the F1 score of a binary classification problem.

  • class FactorScheduler extends LRScheduler

    Class for reducing learning rate in factor

    Class for reducing learning rate in factor

    Assume the weight has been updated by n times, then the learning rate will be base_lr * factor^^(floor(n/step))

  • class FeedForward extends NativeResource

    Model class of MXNet for training and predicting feedforward nets.

    Model class of MXNet for training and predicting feedforward nets. This class is designed for a single-data single output supervised network.

  • abstract class Initializer extends AnyRef

    Base class for Initializer.

  • class KVStore extends NativeResource

  • abstract class LRScheduler extends AnyRef

    Learning rate scheduler, which adaptively changes the learning rate based on the training progress.

  • class MAE extends EvalMetric

    Calculate Mean Absolute Error loss

  • class MSE extends EvalMetric

  • class MXIndexedRecordIO extends MXRecordIO

    Scala interface for read/write RecordIO data formmat with index.

    Scala interface for read/write RecordIO data formmat with index. Support random access.

  • trait MXKVStoreCachedStates extends AnyRef

  • trait MXKVStoreUpdater extends NativeResource

  • class MXNetError extends Exception

  • class MXRecordIO extends AnyRef

    Scala interface for read/write RecordIO data format

  • class Mixed extends Initializer

    Initialize the weight with mixed Initializer

  • class Model extends AnyRef

    Describe the model flow

  • class Monitor extends AnyRef

    Monitor outputs, weights, and gradients for debugging.

  • class NDArray extends NativeResource

    NDArray object in mxnet.

    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.

  • abstract class NDArrayAPIBase extends AnyRef

  • abstract class NDArrayBase extends AnyRef

  • class NDArrayOrScalar[T] extends AnyRef

  • abstract class NDArrayRandomAPIBase extends AnyRef

  • class NameManager extends AnyRef

    NameManager to do automatic naming.

    NameManager to do automatic naming. User can also inherit this object to change naming behavior.

  • class Normal extends Initializer

    Initialize the weight with normal(0, sigma)

  • class NumpyScope extends AnyRef

  • abstract class Optimizer extends Serializable

  • class Perplexity extends EvalMetric

    Calculate perplexity.

  • class RMSE extends EvalMetric

    Calculate Root Mean Squred Error loss

  • class ResourceScope extends AutoCloseable

    This class manages automatically releasing of org.apache.mxnet.NativeResources

  • class Rtc extends AnyRef

    This class allow you to write cuda kernel in Scala and call them with NDArray.

  • class Shape extends Serializable

    Shape of NDArray or other data

  • class Symbol extends NativeResource

    Symbolic configuration API of mxnet.

    Symbolic configuration API of mxnet.
    WARNING: it is your responsibility to clear this object through dispose().

  • abstract class SymbolAPIBase extends AnyRef

  • abstract class SymbolBase extends AnyRef

  • class SymbolConversions[V] extends AnyRef

  • trait SymbolGenerator extends AnyRef

  • class SymbolOrScalar[T] extends AnyRef

  • abstract class SymbolRandomAPIBase extends AnyRef

  • class TestUtil extends AnyRef

  • class TopKAccuracy extends EvalMetric

    Calculate top k predictions accuracy

  • class Uniform extends Initializer

    Initialize the weight with uniform [-scale, scale]

  • trait WarnIfNotDisposed extends AnyRef

    Attributes
    protected

  • class Xavier extends Initializer

    Initialize the weight with Xavier or similar initialization scheme.

  • class NDArrayCollector extends AnyRef

    Annotations
    @deprecated
    Deprecated

    (Since version 1.5.0) Please use ResourceScope instead

    • Value Members

    1. object Callback

      Callback functions that can be used to track various status during epoch.

    2. object CheckUtils

    3. object Context extends Serializable

    4. object DType extends Enumeration

    5. object DataBatch

    6. object DataDesc extends Serializable

    7. object Executor

    8. object FeedForward

    9. object IO

      IO iterators for loading training & validation data

    10. object Image

      Image API of Scala package enable OpenCV feature

    11. object KVStore

      Key value store interface of MXNet for parameter synchronization.

    12. object KVStoreServer

    13. object Layout

      Layout definition of DataDesc N Batch size C channels H Height W Weight T sequence length undefined default value of Layout

    14. object MXRecordIO

    15. object MX_PRIMITIVES

    16. object Model

    17. object NDArray extends NDArrayBase

      NDArray Object extends from NDArrayBase for abstract function signatures Main code will be generated during compile time through Macros

      NDArray Object extends from NDArrayBase for abstract function signatures Main code will be generated during compile time through Macros

      Annotations
      @AddNDArrayFunctions( false )

    18. object NDArrayAPI extends NDArrayAPIBase

      typesafe NDArray API: NDArray.api._ Main code will be generated during compile time through Macros

      typesafe NDArray API: NDArray.api._ Main code will be generated during compile time through Macros

      Annotations
      @AddNDArrayAPIs( false )

    19. object NDArrayOrScalar

    20. object NDArrayRandomAPI extends NDArrayRandomAPIBase

      typesafe NDArray random module: NDArray.random._ Main code will be generated during compile time through Macros

      typesafe NDArray random module: NDArray.random._ Main code will be generated during compile time through Macros

      Annotations
      @AddNDArrayRandomAPIs( false )

    21. object NameManager

    22. object NumpyScope

      NumpyScope object provides util functions for turning on/off NumPy compatibility and checking whether NumPy compatibility has been turned on/off.

      NumpyScope object provides util functions for turning on/off NumPy compatibility and checking whether NumPy compatibility has been turned on/off. NumPy compatibility is introduced first to support zero-dim and zero-size tensors as in NumPy.

    23. object Operator

    24. object Optimizer extends Serializable

    25. object Profiler

    26. object Random

      Random Number interface of mxnet.

    27. object ResourceScope

    28. object Shape extends Serializable

    29. object Symbol extends SymbolBase

      Symbol Object extends from SymbolBase for abstract function signatures Main code will be generated during compile time through Macros

      Symbol Object extends from SymbolBase for abstract function signatures Main code will be generated during compile time through Macros

      Annotations
      @AddSymbolFunctions( false )

    30. object SymbolAPI extends SymbolAPIBase

      typesafe Symbol API: Symbol.api._ Main code will be generated during compile time through Macros

      typesafe Symbol API: Symbol.api._ Main code will be generated during compile time through Macros

      Annotations
      @AddSymbolAPIs( false )

    31. object SymbolConversions

    32. object SymbolOrScalar

    33. object SymbolRandomAPI extends SymbolRandomAPIBase

      typesafe Symbol random module: Symbol.random._ Main code will be generated during compile time through Macros

      typesafe Symbol random module: Symbol.random._ Main code will be generated during compile time through Macros

      Annotations
      @AddSymbolRandomAPIs( false )

    34. object Visualization

    35. package annotation

    36. package contrib

    37. package infer

    38. package io

    39. package module

    40. package optimizer

    41. package util

    Deprecated Value Members

    1. object NDArrayCollector

      A collector to store NDArrays.

      A collector to store NDArrays. It provides a scope, NDArrays allocated in the scope can either

      • be disposed automatically when the code block finishes, or
      • simply be collected for future usage.
        If the return type of scope is NDArray or NDArrayFuncReturn, the collector is smart enough NOT to collect or dispose the returned NDArray.
        However in other cases, it is users' responsibility NOT to leak allocated NDArrays outside, (e.g., store to a global variable and use later, pass to another thread, etc.)
        Usage Example: 
         val a = NDArray.array(Array(-1f, 0f, 1f, 2f, 3f, 4f), shape = Shape(2, 3))
 val res = NDArrayCollector.auto().withScope {
   (NDArray.relu(a) + a).toArray
 }
        In the case above, the intermediate NDArrays (created by NDArray.relu and +) will be disposed automatically.
        User can also decide to dispose the collected NDArrays later: 
         val collector = NDArrayCollector.manual()
 val res = collector.withScope {
   (NDArray.relu(a) + a).toArray
 }
 collector.foreach(_.dispose())
        For Java users: 
         NDArray a = NDArray.array(new float[]{-1f, 0f, 1f, 2f, 3f, 4f},
                           Shape.create(2, 3), Context.cpu(0));
 float[] sliced = NDArrayCollector.auto().withScope(
   new scala.runtime.AbstractFunction0() {
   @Override
   public float[] apply() {
     a.slice(0, 1).toArray();
   }
 });
      Annotations
      @deprecated
      Deprecated

      (Since version 1.5.0) Please use ResourceScope instead

    Ungrouped