ResourceApplyRmsProp
Organízate con las colecciones
Guarda y clasifica el contenido según tus preferencias.
Actualice '*var' según el algoritmo RMSProp.
Tenga en cuenta que en una implementación densa de este algoritmo, ms y mom se actualizarán incluso si el grad es cero, pero en esta implementación escasa, ms y mom no se actualizarán en iteraciones durante las cuales el grad sea cero.
cuadrado_medio = decaimiento * cuadrado_medio + (1-decaimiento) * gradiente ** 2 Delta = tasa_de_aprendizaje * gradiente / sqrt (cuadrado_medio + épsilon)
ms <- rho * ms_{t-1} + (1-rho) * grad * grad mamá <- impulso * mamá_{t-1} + lr * grad / sqrt(ms + épsilon) var <- var - mamá
Constantes
Cadena | OP_NOMBRE | El nombre de esta operación, como lo conoce el motor central de TensorFlow. |
Métodos heredados
De la clase java.lang.Object booleano | es igual (Objeto arg0) |
Clase final<?> | obtenerclase () |
En t | código hash () |
vacío final | notificar () |
vacío final | notificar a todos () |
Cadena | Encadenar () |
vacío final | esperar (arg0 largo, int arg1) |
vacío final | espera (largo arg0) |
vacío final | esperar () |
Constantes
Cadena final estática pública OP_NAME
El nombre de esta operación, como lo conoce el motor central de TensorFlow.
Valor constante: "ResourceApplyRMSProp"
Métodos públicos
Método de fábrica para crear una clase que envuelve una nueva operación ResourceApplyRmsProp.
Parámetros
alcance | alcance actual |
---|
var | Debe ser de una Variable(). |
---|
EM | Debe ser de una Variable(). |
---|
mamá | Debe ser de una Variable(). |
---|
lr | Factor de escala. Debe ser un escalar. |
---|
rho | Tasa de descomposición. Debe ser un escalar. |
---|
épsilon | Término de cresta. Debe ser un escalar. |
---|
graduado | El gradiente. |
---|
opciones | lleva valores de atributos opcionales |
---|
Devoluciones
- una nueva instancia de ResourceApplyRmsProp
Parámetros
utilizarBloqueo | Si es "True", la actualización de los tensores var, ms y mom está protegida por un bloqueo; de lo contrario, el comportamiento no está definido, pero puede presentar menos contención. |
---|
A menos que se indique lo contrario, el contenido de esta página está sujeto a la licencia Reconocimiento 4.0 de Creative Commons y las muestras de código están sujetas a la licencia Apache 2.0. Para obtener más información, consulta las políticas del sitio web de Google Developers. Java es una marca registrada de Oracle o sus afiliados.
Última actualización: 2025-07-26 (UTC).
[null,null,["Última actualización: 2025-07-26 (UTC)."],[],[],null,["# ResourceApplyRmsProp\n\npublic final class **ResourceApplyRmsProp** \nUpdate '\\*var' according to the RMSProp algorithm.\n\n\nNote that in dense implementation of this algorithm, ms and mom will\nupdate even if the grad is zero, but in this sparse implementation, ms\nand mom will not update in iterations during which the grad is zero.\n\n\nmean_square = decay \\* mean_square + (1-decay) \\* gradient \\*\\* 2\nDelta = learning_rate \\* gradient / sqrt(mean_square + epsilon)\n\n\nms \\\u003c- rho \\* ms_{t-1} + (1-rho) \\* grad \\* grad\nmom \\\u003c- momentum \\* mom_{t-1} + lr \\* grad / sqrt(ms + epsilon)\nvar \\\u003c- var - mom\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n### Nested Classes\n\n|-------|---|---|-----------------------------------------------------------------------------------------------------------------|\n| class | [ResourceApplyRmsProp.Options](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp.Options) || Optional attributes for [ResourceApplyRmsProp](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp) |\n\n### Constants\n\n|--------|------------------------------------------------------------------------------------|---------------------------------------------------------|\n| String | [OP_NAME](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp#OP_NAME) | The name of this op, as known by TensorFlow core engine |\n\n### Public Methods\n\n|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| static \\\u003cT extends [TType](/jvm/api_docs/java/org/tensorflow/types/family/TType)\\\u003e [ResourceApplyRmsProp](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp) | [create](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp#create(org.tensorflow.op.Scope, org.tensorflow.Operand\u003c?\u003e, org.tensorflow.Operand\u003c?\u003e, org.tensorflow.Operand\u003c?\u003e, org.tensorflow.Operand\u003cT\u003e, org.tensorflow.Operand\u003cT\u003e, org.tensorflow.Operand\u003cT\u003e, org.tensorflow.Operand\u003cT\u003e, org.tensorflow.Operand\u003cT\u003e, org.tensorflow.op.train.ResourceApplyRmsProp.Options...))([Scope](/jvm/api_docs/java/org/tensorflow/op/Scope) scope, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e var, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e ms, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e mom, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e lr, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e rho, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e momentum, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e epsilon, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e grad, [Options...](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp.Options) options) Factory method to create a class wrapping a new ResourceApplyRmsProp operation. |\n| static [ResourceApplyRmsProp.Options](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp.Options) | [useLocking](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp#useLocking(java.lang.Boolean))(Boolean useLocking) |\n\n### Inherited Methods\n\nFrom class [org.tensorflow.op.RawOp](/jvm/api_docs/java/org/tensorflow/op/RawOp) \n\n|----------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| final boolean | [equals](/jvm/api_docs/java/org/tensorflow/op/RawOp#equals(java.lang.Object))(Object obj) |\n| final int | [hashCode](/jvm/api_docs/java/org/tensorflow/op/RawOp#hashCode())() |\n| [Operation](/jvm/api_docs/java/org/tensorflow/Operation) | [op](/jvm/api_docs/java/org/tensorflow/op/RawOp#op())() Return this unit of computation as a single [Operation](/jvm/api_docs/java/org/tensorflow/Operation). |\n| final String | [toString](/jvm/api_docs/java/org/tensorflow/op/RawOp#toString())() |\n\nFrom class java.lang.Object \n\n|------------------|---------------------------|\n| boolean | equals(Object arg0) |\n| final Class\\\u003c?\\\u003e | getClass() |\n| int | hashCode() |\n| final void | notify() |\n| final void | notifyAll() |\n| String | toString() |\n| final void | wait(long arg0, int arg1) |\n| final void | wait(long arg0) |\n| final void | wait() |\n\nFrom interface [org.tensorflow.op.Op](/jvm/api_docs/java/org/tensorflow/op/Op) \n\n|-----------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| abstract [ExecutionEnvironment](/jvm/api_docs/java/org/tensorflow/ExecutionEnvironment) | [env](/jvm/api_docs/java/org/tensorflow/op/Op#env())() Return the execution environment this op was created in. |\n| abstract [Operation](/jvm/api_docs/java/org/tensorflow/Operation) | [op](/jvm/api_docs/java/org/tensorflow/op/Op#op())() Return this unit of computation as a single [Operation](/jvm/api_docs/java/org/tensorflow/Operation). |\n\nConstants\n---------\n\n#### public static final String\n**OP_NAME**\n\nThe name of this op, as known by TensorFlow core engine \nConstant Value: \"ResourceApplyRMSProp\"\n\nPublic Methods\n--------------\n\n#### public static [ResourceApplyRmsProp](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp)\n**create**\n([Scope](/jvm/api_docs/java/org/tensorflow/op/Scope) scope, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e var, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e ms, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003c?\\\u003e mom, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e lr, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e rho, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e momentum, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e epsilon, [Operand](/jvm/api_docs/java/org/tensorflow/Operand)\\\u003cT\\\u003e grad, [Options...](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp.Options) options)\n\nFactory method to create a class wrapping a new ResourceApplyRmsProp operation. \n\n##### Parameters\n\n| scope | current scope |\n| var | Should be from a Variable(). |\n| ms | Should be from a Variable(). |\n| mom | Should be from a Variable(). |\n| lr | Scaling factor. Must be a scalar. |\n| rho | Decay rate. Must be a scalar. |\n| epsilon | Ridge term. Must be a scalar. |\n| grad | The gradient. |\n| options | carries optional attributes values |\n|---------|------------------------------------|\n\n##### Returns\n\n- a new instance of ResourceApplyRmsProp \n\n#### public static [ResourceApplyRmsProp.Options](/jvm/api_docs/java/org/tensorflow/op/train/ResourceApplyRmsProp.Options)\n**useLocking**\n(Boolean useLocking)\n\n\u003cbr /\u003e\n\n##### Parameters\n\n| useLocking | If \\`True\\`, updating of the var, ms, and mom tensors is protected by a lock; otherwise the behavior is undefined, but may exhibit less contention. |\n|------------|-----------------------------------------------------------------------------------------------------------------------------------------------------|"]]