컬렉션을 사용해 정리하기
내 환경설정을 기준으로 콘텐츠를 저장하고 분류하세요.
텐서플로우:: 작전:: 부분 평균 풀
#include <nn_ops.h>
입력에 대해 부분 평균 풀링을 수행합니다.
요약
부분 평균 풀링은 풀링 영역 생성 단계의 부분 최대 풀링과 유사합니다. 유일한 차이점은 풀링 영역이 생성된 후 각 풀링 영역에서 max 연산 대신 평균 연산이 수행된다는 점입니다.
인수:
- 범위: 범위 개체
- 값:
[batch, height, width, channels]
모양의 4차원. - pooling_ratio:
value
의 각 차원에 대한 풀링 비율. 현재 행 및 열 차원만 지원하며 1.0보다 커야 합니다. 예를 들어 유효한 풀링 비율은 [1.0, 1.44, 1.73, 1.0]과 같습니다. 배치 및 채널 차원에서 풀링을 허용하지 않으므로 첫 번째 요소와 마지막 요소는 1.0이어야 합니다. 1.44와 1.73은 각각 높이와 너비 차원의 풀링 비율입니다.
선택적 속성( Attrs
참조):
- pseudo_random: True로 설정하면 의사 무작위 방식으로 풀링 시퀀스를 생성하고, 그렇지 않으면 무작위 방식으로 생성합니다. 의사 난수와 무작위의 차이에 대해서는 Benjamin Graham의 Fractional Max-Pooling 논문을 확인하세요.
- Overlapping: True로 설정하면 풀링할 때 인접한 풀링 셀 경계의 값이 두 셀 모두에서 사용된다는 의미입니다. 예를 들어:
index 0 1 2 3 4
value 20 5 16 3 7
풀링 시퀀스가 [0, 2, 4]이면 인덱스 2의 16이 두 번 사용됩니다. 부분 평균 풀링의 경우 결과는 [41/3, 26/3]입니다.
- 결정적: True로 설정하면 계산 그래프에서 FractionalAvgPool 노드를 반복할 때 고정 풀링 영역이 사용됩니다. FractionalAvgPool을 결정적으로 만들기 위해 단위 테스트에 주로 사용됩니다.
- Seed: Seed 또는 Seed2가 0이 아닌 값으로 설정된 경우 난수 생성기는 지정된 시드에 의해 시드됩니다. 그렇지 않으면 무작위 시드에 의해 시드됩니다.
- Seed2: 시드 충돌을 피하기 위한 두 번째 시드입니다.
보고:
-
Output
출력: 부분 평균 풀링 후 출력 텐서. -
Output
row_pooling_sequence: 행 풀링 시퀀스, 기울기를 계산하는 데 필요합니다. -
Output
col_pooling_sequence: 기울기를 계산하는 데 필요한 열 풀링 시퀀스입니다.
공개 속성
공공 기능
공개 정적 함수
결정적
Attrs Deterministic(
bool x
)
겹치는
Attrs Overlapping(
bool x
)
의사랜덤
Attrs PseudoRandom(
bool x
)
시드2
Attrs Seed2(
int64 x
)
달리 명시되지 않는 한 이 페이지의 콘텐츠에는 Creative Commons Attribution 4.0 라이선스에 따라 라이선스가 부여되며, 코드 샘플에는 Apache 2.0 라이선스에 따라 라이선스가 부여됩니다. 자세한 내용은 Google Developers 사이트 정책을 참조하세요. 자바는 Oracle 및/또는 Oracle 계열사의 등록 상표입니다.
최종 업데이트: 2025-07-26(UTC)
[null,null,["최종 업데이트: 2025-07-26(UTC)"],[],[],null,["# tensorflow::ops::FractionalAvgPool Class Reference\n\ntensorflow::ops::FractionalAvgPool\n==================================\n\n`#include \u003cnn_ops.h\u003e`\n\nPerforms fractional average pooling on the input.\n\nSummary\n-------\n\nFractional average pooling is similar to Fractional max pooling in the pooling region generation step. The only difference is that after pooling regions are generated, a mean operation is performed instead of a max operation in each pooling region.\n\nArguments:\n\n- scope: A [Scope](/versions/r2.0/api_docs/cc/class/tensorflow/scope#classtensorflow_1_1_scope) object\n- value: 4-D with shape `[batch, height, width, channels]`.\n- pooling_ratio: Pooling ratio for each dimension of `value`, currently only supports row and col dimension and should be \\\u003e= 1.0. For example, a valid pooling ratio looks like \\[1.0, 1.44, 1.73, 1.0\\]. The first and last elements must be 1.0 because we don't allow pooling on batch and channels dimensions. 1.44 and 1.73 are pooling ratio on height and width dimensions respectively.\n\n\u003cbr /\u003e\n\nOptional attributes (see [Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs#structtensorflow_1_1ops_1_1_fractional_avg_pool_1_1_attrs)):\n\n- pseudo_random: When set to True, generates the pooling sequence in a pseudorandom fashion, otherwise, in a random fashion. Check paper [Benjamin Graham, Fractional Max-Pooling](http://arxiv.org/abs/1412.6071) for difference between pseudorandom and random.\n- overlapping: When set to True, it means when pooling, the values at the boundary of adjacent pooling cells are used by both cells. For example:\n\n\u003cbr /\u003e\n\n\n`index 0 1 2 3 4`\n\n\n`value 20 5 16 3 7`\n\nIf the pooling sequence is \\[0, 2, 4\\], then 16, at index 2 will be used twice. The result would be \\[41/3, 26/3\\] for fractional avg pooling.\n\n- deterministic: When set to True, a fixed pooling region will be used when iterating over a [FractionalAvgPool](/versions/r2.0/api_docs/cc/class/tensorflow/ops/fractional-avg-pool#classtensorflow_1_1ops_1_1_fractional_avg_pool) node in the computation graph. Mainly used in unit test to make [FractionalAvgPool](/versions/r2.0/api_docs/cc/class/tensorflow/ops/fractional-avg-pool#classtensorflow_1_1ops_1_1_fractional_avg_pool) deterministic.\n- seed: If either seed or seed2 are set to be non-zero, the random number generator is seeded by the given seed. Otherwise, it is seeded by a random seed.\n- seed2: An second seed to avoid seed collision.\n\n\u003cbr /\u003e\n\nReturns:\n\n- [Output](/versions/r2.0/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output) output: output tensor after fractional avg pooling.\n- [Output](/versions/r2.0/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output) row_pooling_sequence: row pooling sequence, needed to calculate gradient.\n- [Output](/versions/r2.0/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output) col_pooling_sequence: column pooling sequence, needed to calculate gradient.\n\n\u003cbr /\u003e\n\n| ### Constructors and Destructors ||\n|---|---|\n| [FractionalAvgPool](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1a83af6f6e93dbac2bf42ad6afc05d2a86)`(const ::`[tensorflow::Scope](/versions/r2.0/api_docs/cc/class/tensorflow/scope#classtensorflow_1_1_scope)` & scope, ::`[tensorflow::Input](/versions/r2.0/api_docs/cc/class/tensorflow/input#classtensorflow_1_1_input)` value, const gtl::ArraySlice\u003c float \u003e & pooling_ratio)` ||\n| [FractionalAvgPool](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1afe59c1134290e6cfe190960e53e836ed)`(const ::`[tensorflow::Scope](/versions/r2.0/api_docs/cc/class/tensorflow/scope#classtensorflow_1_1_scope)` & scope, ::`[tensorflow::Input](/versions/r2.0/api_docs/cc/class/tensorflow/input#classtensorflow_1_1_input)` value, const gtl::ArraySlice\u003c float \u003e & pooling_ratio, const `[FractionalAvgPool::Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs#structtensorflow_1_1ops_1_1_fractional_avg_pool_1_1_attrs)` & attrs)` ||\n\n| ### Public attributes ||\n|------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------|\n| [col_pooling_sequence](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1a253a9b7940b383f04c70aa5254f52995) | `::`[tensorflow::Output](/versions/r2.0/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output) |\n| [operation](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1a8b1bbb7c981afe922b39753597ab754b) | [Operation](/versions/r2.0/api_docs/cc/class/tensorflow/operation#classtensorflow_1_1_operation) |\n| [output](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1a72c1fe35152d17096cfcd5ca3d626e24) | `::`[tensorflow::Output](/versions/r2.0/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output) |\n| [row_pooling_sequence](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1aef40ec50b456803bb75a8474cdc29fcb) | `::`[tensorflow::Output](/versions/r2.0/api_docs/cc/class/tensorflow/output#classtensorflow_1_1_output) |\n\n| ### Public static functions ||\n|---------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------|\n| [Deterministic](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1a286c7e7d0ea4b667eb0fca780f6c8fd8)`(bool x)` | [Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs#structtensorflow_1_1ops_1_1_fractional_avg_pool_1_1_attrs) |\n| [Overlapping](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1a561400c14f7e0877122cf0faad67b785)`(bool x)` | [Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs#structtensorflow_1_1ops_1_1_fractional_avg_pool_1_1_attrs) |\n| [PseudoRandom](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1aaeb0a37c716692070fa056b6f164adab)`(bool x)` | [Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs#structtensorflow_1_1ops_1_1_fractional_avg_pool_1_1_attrs) |\n| [Seed](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1a691079eab5c004dc817e928c12380fe5)`(int64 x)` | [Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs#structtensorflow_1_1ops_1_1_fractional_avg_pool_1_1_attrs) |\n| [Seed2](#classtensorflow_1_1ops_1_1_fractional_avg_pool_1aba6caf6e7f50e68e728b8ac9357b9353)`(int64 x)` | [Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs#structtensorflow_1_1ops_1_1_fractional_avg_pool_1_1_attrs) |\n\n| ### Structs ||\n|-------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| [tensorflow::ops::FractionalAvgPool::Attrs](/versions/r2.0/api_docs/cc/struct/tensorflow/ops/fractional-avg-pool/attrs) | Optional attribute setters for [FractionalAvgPool](/versions/r2.0/api_docs/cc/class/tensorflow/ops/fractional-avg-pool#classtensorflow_1_1ops_1_1_fractional_avg_pool). |\n\nPublic attributes\n-----------------\n\n### col_pooling_sequence\n\n```scdoc\n::tensorflow::Output col_pooling_sequence\n``` \n\n### operation\n\n```text\nOperation operation\n``` \n\n### output\n\n```text\n::tensorflow::Output output\n``` \n\n### row_pooling_sequence\n\n```scdoc\n::tensorflow::Output row_pooling_sequence\n``` \n\nPublic functions\n----------------\n\n### FractionalAvgPool\n\n```gdscript\n FractionalAvgPool(\n const ::tensorflow::Scope & scope,\n ::tensorflow::Input value,\n const gtl::ArraySlice\u003c float \u003e & pooling_ratio\n)\n``` \n\n### FractionalAvgPool\n\n```gdscript\n FractionalAvgPool(\n const ::tensorflow::Scope & scope,\n ::tensorflow::Input value,\n const gtl::ArraySlice\u003c float \u003e & pooling_ratio,\n const FractionalAvgPool::Attrs & attrs\n)\n``` \n\nPublic static functions\n-----------------------\n\n### Deterministic\n\n```text\nAttrs Deterministic(\n bool x\n)\n``` \n\n### Overlapping\n\n```text\nAttrs Overlapping(\n bool x\n)\n``` \n\n### PseudoRandom\n\n```text\nAttrs PseudoRandom(\n bool x\n)\n``` \n\n### Seed\n\n```text\nAttrs Seed(\n int64 x\n)\n``` \n\n### Seed2\n\n```text\nAttrs Seed2(\n int64 x\n)\n```"]]