Organízate con las colecciones
Guarda y clasifica el contenido según tus preferencias.
flujo tensor:: operaciones:: Lote al espacioND
#include <array_ops.h> BatchToSpace para tensores ND de tipo T.
Resumen
Esta operación reforma la dimensión "lote" 0 en M + 1 dimensiones de forma block_shape + [batch] , entrelaza estos bloques nuevamente en la cuadrícula definida por las dimensiones espaciales [1, ..., M] , para obtener un resultado con el mismo rango que la entrada. Las dimensiones espaciales de este resultado intermedio se recortan opcionalmente según crops para producir el resultado. Esto es lo contrario de SpaceToBatch. Consulte a continuación para obtener una descripción precisa.
Argumentos:
- alcance: un objeto de alcance
- entrada: ND con forma
input_shape = [batch] + spatial_shape + remaining_shape , donde forma_espacial tiene M dimensiones. - block_shape: 1-D con forma
[M] , todos los valores deben ser >= 1. - cultivos: 2-D con forma
[M, 2] , todos los valores deben ser >= 0. crops[i] = [crop_start, crop_end] especifica la cantidad a recortar de la dimensión de entrada i + 1 , que corresponde a la dimensión espacial i . Se requiere que crop_start[i] + crop_end[i] <= block_shape[i] * input_shape[i + 1] .
Esta operación equivale a los siguientes pasos:
- Transformar
input para reshaped la forma: [block_shape[0], ..., block_shape[M-1], lote / prod(block_shape), input_shape[1], ..., input_shape[N-1]] - Permutar dimensiones de
reshaped para producir forma permuted [batch / prod(block_shape),input_shape[1], block_shape[0], ..., input_shape[M], block_shape[M-1],input_shape[M+1], ..., forma_entrada[N-1]] - Reforma
permuted para producir reshaped_permuted de forma [batch / prod(block_shape),input_shape[1] * block_shape[0], ..., input_shape[M] * block_shape[M-1],input_shape[M+1], .. ., forma_entrada[N-1]] - Recorte el inicio y el final de las dimensiones
[1, ..., M] de reshaped_permuted según los crops para producir la salida de forma: [batch / prod(block_shape),input_shape[1] * block_shape[0] - crop[0, 0] - cultivos[0,1], ..., forma_entrada[M] * forma_bloque[M-1] - cultivos[M-1,0] - cultivos[M-1,1],forma_entrada[M+1] , ..., forma_entrada[N-1]]
Algunos ejemplos:
(1) Para la siguiente entrada de forma [4, 1, 1, 1] , block_shape = [2, 2] y crops = [[0, 0], [0, 0]] :
[[[[1]]], [[[2]]], [[[3]]], [[[4]]]]
El tensor de salida tiene forma [1, 2, 2, 1] y valor:
x = [[[[1], [2]], [[3], [4]]]]
(2) Para la siguiente entrada de forma [4, 1, 1, 3] , block_shape = [2, 2] y crops = [[0, 0], [0, 0]] :
[[[[1, 2, 3]]], [[[4, 5, 6]]], [[[7, 8, 9]]], [[[10, 11, 12]]]]
El tensor de salida tiene forma [1, 2, 2, 3] y valor:
x = [[[[1, 2, 3], [4, 5, 6]],
[[7, 8, 9], [10, 11, 12]]]] (3) Para la siguiente entrada de forma [4, 2, 2, 1] , block_shape = [2, 2] y crops = [[0, 0], [0, 0]] :
x = [[[[1], [3]], [[9], [11]]],
[[[2], [4]], [[10], [12]]],
[[[5], [7]], [[13], [15]]],
[[[6], [8]], [[14], [16]]]] El tensor de salida tiene forma [1, 4, 4, 1] y valor:
x = [[[[1], [2], [3], [4]],
[[5], [6], [7], [8]],
[[9], [10], [11], [12]],
[[13], [14], [15], [16]]]] (4) Para la siguiente entrada de forma [8, 1, 3, 1] , block_shape = [2, 2] y crops = [[0, 0], [2, 0]] :
x = [[[[0], [1], [3]]], [[[0], [9], [11]]],
[[[0], [2], [4]]], [[[0], [10], [12]]],
[[[0], [5], [7]]], [[[0], [13], [15]]],
[[[0], [6], [8]]], [[[0], [14], [16]]]] El tensor de salida tiene forma [2, 2, 4, 1] y valor:
x = [[[[1], [2], [3], [4]],
[[5], [6], [7], [8]]],
[[[9], [10], [11], [12]],
[[13], [14], [15], [16]]]] Devoluciones:
Atributos públicos
Funciones públicas
nodo
::tensorflow::Node * node() const
operator::tensorflow::Input() const
operador::tensorflow::Salida
operator::tensorflow::Output() const
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,["# tensorflow::ops::BatchToSpaceND Class Reference\n\ntensorflow::ops::BatchToSpaceND\n===============================\n\n`#include \u003carray_ops.h\u003e`\n\n[BatchToSpace](/versions/r1.15/api_docs/cc/class/tensorflow/ops/batch-to-space#classtensorflow_1_1ops_1_1_batch_to_space) for N-D tensors of type T.\n\nSummary\n-------\n\nThis operation reshapes the \"batch\" dimension 0 into `M + 1` dimensions of shape `block_shape + [batch]`, interleaves these blocks back into the grid defined by the spatial dimensions `[1, ..., M]`, to obtain a result with the same rank as the input. The spatial dimensions of this intermediate result are then optionally cropped according to `crops` to produce the output. This is the reverse of SpaceToBatch. See below for a precise description.\n\nArguments:\n\n- scope: A [Scope](/versions/r1.15/api_docs/cc/class/tensorflow/scope#classtensorflow_1_1_scope) object\n- input: N-D with shape `input_shape = [batch] + spatial_shape + remaining_shape`, where spatial_shape has M dimensions.\n- block_shape: 1-D with shape `[M]`, all values must be \\\u003e= 1.\n- crops: 2-D with shape `[M, 2]`, all values must be \\\u003e= 0. `crops[i] = [crop_start, crop_end]` specifies the amount to crop from input dimension `i + 1`, which corresponds to spatial dimension `i`. It is required that `crop_start[i] + crop_end[i] \u003c= block_shape[i] * input_shape[i + 1]`.\n\n\u003cbr /\u003e\n\nThis operation is equivalent to the following steps:\n\n\n1. Reshape `input` to `reshaped` of shape: \\[block_shape\\[0\\], ..., block_shape\\[M-1\\], batch / prod(block_shape), input_shape\\[1\\], ..., input_shape\\[N-1\\]\\]\n2. Permute dimensions of `reshaped` to produce `permuted` of shape \\[batch / prod(block_shape),input_shape\\[1\\], block_shape\\[0\\], ..., input_shape\\[M\\], block_shape\\[M-1\\],input_shape\\[M+1\\], ..., input_shape\\[N-1\\]\\]\n3. Reshape `permuted` to produce `reshaped_permuted` of shape \\[batch / prod(block_shape),input_shape\\[1\\] \\* block_shape\\[0\\], ..., input_shape\\[M\\] \\* block_shape\\[M-1\\],input_shape\\[M+1\\], ..., input_shape\\[N-1\\]\\]\n4. Crop the start and end of dimensions `[1, ..., M]` of `reshaped_permuted` according to `crops` to produce the output of shape: \\[batch / prod(block_shape),input_shape\\[1\\] \\* block_shape\\[0\\] - crops\\[0,0\\] - crops\\[0,1\\], ..., input_shape\\[M\\] \\* block_shape\\[M-1\\] - crops\\[M-1,0\\] - crops\\[M-1,1\\],input_shape\\[M+1\\], ..., input_shape\\[N-1\\]\\]\n\n\u003cbr /\u003e\n\nSome examples:\n\n(1) For the following input of shape `[4, 1, 1, 1]`, `block_shape = [2, 2]`, and `crops = [[0, 0], [0, 0]]`:\n\n\n```text\n[[[[1]]], [[[2]]], [[[3]]], [[[4]]]]\n```\n\n\u003cbr /\u003e\n\nThe output tensor has shape `[1, 2, 2, 1]` and value:\n\n\n```text\nx = [[[[1], [2]], [[3], [4]]]]\n```\n\n\u003cbr /\u003e\n\n(2) For the following input of shape `[4, 1, 1, 3]`, `block_shape = [2, 2]`, and `crops = [[0, 0], [0, 0]]`:\n\n\n```text\n[[[[1, 2, 3]]], [[[4, 5, 6]]], [[[7, 8, 9]]], [[[10, 11, 12]]]]\n```\n\n\u003cbr /\u003e\n\nThe output tensor has shape `[1, 2, 2, 3]` and value:\n\n\n```text\nx = [[[[1, 2, 3], [4, 5, 6]],\n [[7, 8, 9], [10, 11, 12]]]]\n```\n\n\u003cbr /\u003e\n\n(3) For the following input of shape `[4, 2, 2, 1]`, `block_shape = [2, 2]`, and `crops = [[0, 0], [0, 0]]`:\n\n\n```text\nx = [[[[1], [3]], [[9], [11]]],\n [[[2], [4]], [[10], [12]]],\n [[[5], [7]], [[13], [15]]],\n [[[6], [8]], [[14], [16]]]]\n```\n\n\u003cbr /\u003e\n\nThe output tensor has shape `[1, 4, 4, 1]` and value:\n\n\n```text\nx = [[[[1], [2], [3], [4]],\n [[5], [6], [7], [8]],\n [[9], [10], [11], [12]],\n [[13], [14], [15], [16]]]]\n```\n\n\u003cbr /\u003e\n\n(4) For the following input of shape `[8, 1, 3, 1]`, `block_shape = [2, 2]`, and `crops = [[0, 0], [2, 0]]`:\n\n\n```text\nx = [[[[0], [1], [3]]], [[[0], [9], [11]]],\n [[[0], [2], [4]]], [[[0], [10], [12]]],\n [[[0], [5], [7]]], [[[0], [13], [15]]],\n [[[0], [6], [8]]], [[[0], [14], [16]]]]\n```\n\n\u003cbr /\u003e\n\nThe output tensor has shape `[2, 2, 4, 1]` and value:\n\n\n```text\nx = [[[[1], [2], [3], [4]],\n [[5], [6], [7], [8]]],\n [[[9], [10], [11], [12]],\n [[13], [14], [15], [16]]]]\n```\n\n\u003cbr /\u003e\n\nReturns:\n\n- [Output](/versions/r1.15/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output): The output tensor.\n\n\u003cbr /\u003e\n\n| ### Constructors and Destructors ||\n|---|---|\n| [BatchToSpaceND](#classtensorflow_1_1ops_1_1_batch_to_space_n_d_1ae9fc7cf839b67ec1692eb9dbd13dab3f)`(const ::`[tensorflow::Scope](/versions/r1.15/api_docs/cc/class/tensorflow/scope#classtensorflow_1_1_scope)` & scope, ::`[tensorflow::Input](/versions/r1.15/api_docs/cc/class/tensorflow/input#classtensorflow_1_1_input)` input, ::`[tensorflow::Input](/versions/r1.15/api_docs/cc/class/tensorflow/input#classtensorflow_1_1_input)` block_shape, ::`[tensorflow::Input](/versions/r1.15/api_docs/cc/class/tensorflow/input#classtensorflow_1_1_input)` crops)` ||\n\n| ### Public attributes ||\n|------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------|\n| [operation](#classtensorflow_1_1ops_1_1_batch_to_space_n_d_1a1e8d19aed27a8ba75041200ee25a7310) | [Operation](/versions/r1.15/api_docs/cc/class/tensorflow/operation#classtensorflow_1_1_operation) |\n| [output](#classtensorflow_1_1ops_1_1_batch_to_space_n_d_1a2f9a5258c2d37ba9ce71c6ebfe2f754d) | `::`[tensorflow::Output](/versions/r1.15/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output) |\n\n| ### Public functions ||\n|------------------------------------------------------------------------------------------------------------------------------|------------------------|\n| [node](#classtensorflow_1_1ops_1_1_batch_to_space_n_d_1a8c320b154abac62302b289161e5aa745)`() const ` | `::tensorflow::Node *` |\n| [operator::tensorflow::Input](#classtensorflow_1_1ops_1_1_batch_to_space_n_d_1a94adde19cfddf4d1109cceff401543c8)`() const ` | ` ` ` ` |\n| [operator::tensorflow::Output](#classtensorflow_1_1ops_1_1_batch_to_space_n_d_1a17e07f190557e6565111355cc159b528)`() const ` | ` ` ` ` |\n\nPublic attributes\n-----------------\n\n### operation\n\n```text\nOperation operation\n``` \n\n### output\n\n```text\n::tensorflow::Output output\n``` \n\nPublic functions\n----------------\n\n### BatchToSpaceND\n\n```gdscript\n BatchToSpaceND(\n const ::tensorflow::Scope & scope,\n ::tensorflow::Input input,\n ::tensorflow::Input block_shape,\n ::tensorflow::Input crops\n)\n``` \n\n### node\n\n```gdscript\n::tensorflow::Node * node() const \n``` \n\n### operator::tensorflow::Input\n\n```gdscript\n operator::tensorflow::Input() const \n``` \n\n### operator::tensorflow::Output\n\n```gdscript\n operator::tensorflow::Output() const \n```"]]
