Conozca lo último en aprendizaje automático, IA generativa y más en el
Simposio WiML 2023.
EagerSession.Options
Organiza tus páginas con colecciones
Guarda y categoriza el contenido según tus preferencias.
Métodos Heredados
De la clase java.lang.Object | booleano | es igual a (Objeto arg0) |
| clase final<?> | obtenerClase () |
| En t | código hash () |
| vacío final | notificar () |
| vacío final | notificar a todos () |
| Cuerda | a la cadena () |
| vacío final | espera (largo arg0, int arg1) |
| vacío final | espera (largo arg0) |
| vacío final | espera () |
Métodos públicos
Controla cómo se ejecutan realmente las operaciones despachadas.
Cuando se establece en verdadero, cada operación se ejecuta de forma asincrónica (en cuyo caso, algunas operaciones pueden devolver salidas "no preparadas"). Cuando se establece en falso, todas las operaciones se ejecutan sincrónicamente.
La ejecución síncrona se utiliza de forma predeterminada.
Parámetros
| valor | true para ejecución asíncrona, false para ejecución síncrona. |
|---|
Construye una sesión ansiosa con las opciones seleccionadas.
Configuración pública de EagerSession.Options (valor de byte [])
Configura la sesión en función de los datos que se encuentran en el búfer proporcionado, que es un protocolo de configuración de TensorFlow serializado.
Advertencia: la compatibilidad con esta función está sujeta a cambios, ya que es posible que los prototipos de TensorFlow no se admitan en terminales públicos en el futuro.
Ver también: config.proto
Parámetros
| valor | un protocolo de configuración serializado |
|---|
Controla cómo actuar cuando intentamos ejecutar una operación en un dispositivo determinado pero algunos tensores de entrada no están en ese dispositivo.
EagerSession.DevicePlacementPolicy.SILENT se usa de forma predeterminada.
Parámetros
| valor | estrategia a utilizar |
|---|
Salvo que se indique lo contrario, el contenido de esta página está sujeto a la licencia Atribución 4.0 de Creative Commons, y los ejemplos de código están sujetos a la licencia Apache 2.0. Para obtener más información, consulta las políticas del sitio de Google Developers. Java es una marca registrada de Oracle o sus afiliados. Cierto contenido está sujeto a la licencia de NumPy.
Última actualización: 2022-05-17 (UTC)
[null,null,["Última actualización: 2022-05-17 (UTC)"],[],[],null,["# EagerSession.Options\n\npublic static class **EagerSession.Options** \n\n### Public Methods\n\n|------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options) | [async](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options#async(boolean))(boolean value) Controls how operations dispatched are actually executed. |\n| [EagerSession](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession) | [build](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options#build())() Builds an eager session with the selected options. |\n| [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options) | [config](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options#config(byte[]))(byte\\[\\] value) Configures the session based on the data found in the provided buffer, which is serialized TensorFlow config proto. |\n| [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options) | [devicePlacementPolicy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options#devicePlacementPolicy(org.tensorflow.EagerSession.DevicePlacementPolicy))([EagerSession.DevicePlacementPolicy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.DevicePlacementPolicy) value) Controls how to act when we try to run an operation on a given device but some input tensors are not on that device. |\n| [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options) | [resourceCleanupStrategy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options#resourceCleanupStrategy(org.tensorflow.EagerSession.ResourceCleanupStrategy))([EagerSession.ResourceCleanupStrategy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.ResourceCleanupStrategy) value) Controls how TensorFlow resources are cleaned up when no longer needed. |\n\n### Inherited Methods\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\nPublic Methods\n--------------\n\n#### public [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options)\n**async**\n(boolean value)\n\nControls how operations dispatched are actually executed.\n\nWhen set to true, each operation are executed asynchronously (in which case some\noperations might return \"non-ready\" outputs). When set to false, all operations are executed\nsynchronously.\n\nSynchronous execution is used by default.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n##### Parameters\n\n| value | true for asynchronous execution, false for synchronous. |\n|-------|---------------------------------------------------------|\n\n#### public [EagerSession](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession)\n**build**\n()\n\nBuilds an eager session with the selected options. \n\n#### public [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options)\n**config**\n(byte\\[\\] value)\n\nConfigures the session based on the data found in the provided buffer, which is serialized\nTensorFlow config proto.\n\nWarning: the support of this feature is subject to changes since TensorFlow protos might\nnot be supported on public endpoints in the future.\n\nSee also: [config.proto](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/protobuf/config.proto)\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n##### Parameters\n\n| value | a serialized config proto |\n|-------|---------------------------|\n\n#### public [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options)\n**devicePlacementPolicy**\n([EagerSession.DevicePlacementPolicy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.DevicePlacementPolicy) value)\n\nControls how to act when we try to run an operation on a given device but some input tensors\nare not on that device.\n\n[EagerSession.DevicePlacementPolicy.SILENT](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.DevicePlacementPolicy#SILENT) is used by default.\n\n\u003cbr /\u003e\n\n##### Parameters\n\n| value | policy to apply |\n|-------|-----------------|\n\n##### See Also\n\n- [EagerSession.DevicePlacementPolicy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.DevicePlacementPolicy) \n\n#### public [EagerSession.Options](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.Options)\n**resourceCleanupStrategy**\n([EagerSession.ResourceCleanupStrategy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.ResourceCleanupStrategy) value)\n\nControls how TensorFlow resources are cleaned up when no longer needed.\n\n[EagerSession.ResourceCleanupStrategy.IN_BACKGROUND](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.ResourceCleanupStrategy#IN_BACKGROUND) is used by default.\n\n\u003cbr /\u003e\n\n##### Parameters\n\n| value | strategy to use |\n|-------|-----------------|\n\n##### See Also\n\n- [EagerSession.ResourceCleanupStrategy](/versions/r2.9/api_docs/java/org/tensorflow/EagerSession.ResourceCleanupStrategy)"]]
