Packages

  • package root
    Definition Classes
    root
  • package com
    Definition Classes
    root
  • package nvidia
    Definition Classes
    com
  • package mdx

    this is the base transform for mdx analytics SDK, the transform comprise of

    this is the base transform for mdx analytics SDK, the transform comprise of


    core : mdx streaming core classes
    xapp : comprise of sample applications

    Definition Classes
    nvidia
  • package core

    this is the core transform for mdx SDK, contains transform and classes for the core mdx functionality

    this is the core transform for mdx SDK, contains transform and classes for the core mdx functionality

    Definition Classes
    mdx
  • package stream
    Definition Classes
    core
  • ObjectState
  • Sink
  • StateMgmt
  • jObjectState
  • jSink
  • jStateMgmt
c

com.nvidia.mdx.core.stream

jStateMgmt

class jStateMgmt extends Serializable

State Management, for object tracking over a period of time, use for metadata in json

// init state management
val stateMgmt = new StateMgmt(config)

//usage for dataset of Message
dataset
  .groupByKey(msg => msg.key)
  .mapGroupsWithState(processingTimeTimeout())(stateMgmt.updateBehavior)
Linear Supertypes
Serializable, Serializable, AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. jStateMgmt
  2. Serializable
  3. Serializable
  4. AnyRef
  5. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new jStateMgmt(config: Map[String, String])

Value Members

  1. final def !=(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  2. final def ##(): Int
    Definition Classes
    AnyRef → Any
  3. final def ==(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  4. final def asInstanceOf[T0]: T0
    Definition Classes
    Any
  5. def clone(): AnyRef
    Attributes
    protected[lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... ) @native() @HotSpotIntrinsicCandidate()
  6. final def eq(arg0: AnyRef): Boolean
    Definition Classes
    AnyRef
  7. def equals(arg0: Any): Boolean
    Definition Classes
    AnyRef → Any
  8. final def getClass(): Class[_]
    Definition Classes
    AnyRef → Any
    Annotations
    @native() @HotSpotIntrinsicCandidate()
  9. def groupBySensorId(id: String, inputs: Iterator[String]): (String, List[String])

    used with simple map group for behavior JSON

    used with simple map group for behavior JSON

     dataset //Behavior JSON dataset
.groupByKey { json =>
       val sensorId = extract sensorId from json
       sensorId
     }.mapGroups(stateMgmt.groupBySensorId)
  10. def hashCode(): Int
    Definition Classes
    AnyRef → Any
    Annotations
    @native() @HotSpotIntrinsicCandidate()
  11. final def isInstanceOf[T0]: Boolean
    Definition Classes
    Any
  12. final def ne(arg0: AnyRef): Boolean
    Definition Classes
    AnyRef
  13. final def notify(): Unit
    Definition Classes
    AnyRef
    Annotations
    @native() @HotSpotIntrinsicCandidate()
  14. final def notifyAll(): Unit
    Definition Classes
    AnyRef
    Annotations
    @native() @HotSpotIntrinsicCandidate()
  15. final def synchronized[T0](arg0: ⇒ T0): T0
    Definition Classes
    AnyRef
  16. def toString(): String
    Definition Classes
    AnyRef → Any
  17. def updateBehavior(objectId: String, inputs: Iterator[jMessage], oldState: GroupState[jObjectState]): Option[jBehavior]

    tracks and maintains state for a given object in GEO spatial coordinates, internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    tracks and maintains state for a given object in GEO spatial coordinates, internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    it does the following functions

    • remove any ObjectState which is timed out.
    • create new ObjectState, if no similar objectId was seen before or existing ObjectId was last seen long time back
    • update existing ObjectState if the new data event timestamp is within configurable time span
    • prune ObjectState as needed, when the object is seen for long period of times, say parked car, the number of points are pruned to reduce the memory footprint

    ObjectState management saves/stores least amount of information to reduce the storage,

    • sensorId + objectId
    • start and end timestamp
    • sequence of geo points
    • object specific attributes are stored as json, example car vs people attributes. Json/string is used to keep the ObjectState portable across different kind of possible object application need to support
    • other attributes like Pose, Gaze, Gesture

    The internal state is translated to Object com.nvidia.mdx.core.schema.jBehavior before sending it to any sink for consumption

    this function is used in conjuction to "apache spark Stateful transformations with mapGroupsWithState"
The mapGroupsWithState(timeoutConf: GroupStateTimeout)(func: (K, Iterator[V], GroupState[S]) => U) is a
transformation applied on a group of data. Since it requires the data to be grouped, it introduces a
big risk of shuffling. The state is computed by calling the parameter called func (in this case updateTrajectory).
This transformation can be used on a bounder and unbounded source.
For this first case, the final state is computed immediately while in the second one the state can
change in every triggered processing. It's the reason why it's stored in a fault-tolerant store
between subsequent invocations.
  18. def updateBehaviorE(objectId: String, inputs: Iterator[jMessage], oldState: GroupState[jObjectState]): Option[jBehavior]

    tracks and maintains state for a given object Euclidean space , internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    tracks and maintains state for a given object Euclidean space , internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    it does the following functions

    • remove any ObjectState which is timed out.
    • create new ObjectState, if no similar objectId was seen before or existing ObjectId was last seen long time back
    • update existing ObjectState if the new data event timestamp is within configurable time span
    • prune ObjectState as needed, when the object is seen for long period of times, say parked car, the number of points are pruned to reduce the memory footprint

    ObjectState state management saves/stores least amount of information to reduce the storage,

    • sensorId + objectId
    • start and end timestamp
    • sequence of geo points
    • object specific attributes are stored as json, example car vs people attributes. Json/string is used to keep the ObjectState portable across different kind of possible object application need to support
    • other attributes like Pose, Gaze, Gesture

    The internal state is translated to Object com.nvidia.mdx.core.schema.jBehavior before sending it to any sink for consumption

    this function is used in conjuction to "apache spark Stateful transformations with mapGroupsWithState"
The mapGroupsWithState(timeoutConf: GroupStateTimeout)(func: (K, Iterator[V], GroupState[S]) => U) is a
transformation applied on a group of data. Since it requires the data to be grouped, it introduces a
big risk of shuffling. The state is computed by calling the parameter called func (in this case updateTrajectory).
This transformation can be used on a bounder and unbounded source.
For this first case, the final state is computed immediately while in the second one the state can
change in every triggered processing. It's the reason why it's stored in a fault-tolerant store
between subsequent invocations.
  19. def updateBehaviorEwithTripwire(objectId: String, inputs: Iterator[jMessage], oldState: GroupState[jObjectState]): Option[(jBehavior, jBehavior)]

    tracks and maintains state for a given object Euclidean space , internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    tracks and maintains state for a given object Euclidean space , internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    it does the following functions

    • remove any ObjectState which is timed out.
    • create new ObjectState, if no similar objectId was seen before or existing ObjectId was last seen long time back
    • update existing ObjectState if the new data event timestamp is within configurable time span
    • prune ObjectState as needed, when the object is seen for long period of times, say parked car, the number of points are pruned to reduce the memory footprint

    ObjectState management saves/stores least amount of information to reduce the storage,

    • sensorId + objectId
    • start and end timestamp
    • sequence of geo points
    • object specific attributes are stored as json, example car vs people attributes. Json/string is used to keep the ObjectState portable across different kind of possible object application need to support
    • other attributes like Pose, Gaze, Gesture

    The internal state is translated to Object com.nvidia.mdx.core.schema.jBehavior before sending it to any sink for consumption. this also generated partial Behavior, basically partial tracklets for further processing of tripwire events

    this function is used in conjuction to "apache spark Stateful transformations with mapGroupsWithState"
The mapGroupsWithState(timeoutConf: GroupStateTimeout)(func: (K, Iterator[V], GroupState[S]) => U) is a
transformation applied on a group of data. Since it requires the data to be grouped, it introduces a
big risk of shuffling. The state is computed by calling the parameter called func (in this case updateTrajectory).
This transformation can be used on a bounder and unbounded source.
For this first case, the final state is computed immediately while in the second one the state can
change in every triggered processing. It's the reason why it's stored in a fault-tolerant store
between subsequent invocations.
  20. def updateBehaviorI(objectId: String, inputs: Iterator[jMessage], oldState: GroupState[jObjectState]): Option[(jBehavior, jBehavior)]

    tracks and maintains state for a given object Image space , internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    tracks and maintains state for a given object Image space , internally named ObjectState, the ID is uniquely identified using sensorId + ObjectId

    it does the following functions

    • remove any ObjectState which is timed out.
    • create new ObjectState, if no similar objectId was seen before or existing ObjectId was last seen long time back
    • update existing ObjectState if the new data event timestamp is within configurable time span
    • prune ObjectState as needed, when the object is seen for long period of times, say parked car, the number of points are pruned to reduce the memory footprint

    ObjectState management saves/stores least amount of information to reduce the storage,

    • sensorId + objectId
    • start and end timestamp
    • sequence of geo points
    • object specific attributes are stored as json, example car vs people attributes. Json/string is used to keep the ObjectState portable across different kind of possible object application need to support
    • other attributes like Pose, Gaze, Gesture

    The internal state is translated to Object com.nvidia.mdx.core.schema.jBehavior before sending it to any sink for consumption. this also generated partial Behavior, basically partial tracklets for further processing of tripwire events

    this function is used in conjuction to "apache spark Stateful transformations with mapGroupsWithState"
The mapGroupsWithState(timeoutConf: GroupStateTimeout)(func: (K, Iterator[V], GroupState[S]) => U) is a
transformation applied on a group of data. Since it requires the data to be grouped, it introduces a
big risk of shuffling. The state is computed by calling the parameter called func (in this case updateTrajectory).
This transformation can be used on a bounder and unbounded source.
For this first case, the final state is computed immediately while in the second one the state can
change in every triggered processing. It's the reason why it's stored in a fault-tolerant store
between subsequent invocations.
  21. final def wait(arg0: Long, arg1: Int): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  22. final def wait(arg0: Long): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... ) @native()
  23. final def wait(): Unit
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )

Deprecated Value Members

  1. def finalize(): Unit
    Attributes
    protected[lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] ) @Deprecated
    Deprecated

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped