FilesLast commitLast update
23 days ago
4 months ago
8 days ago
1 month ago
8 days ago
2 months ago
15 days ago
7 months ago
1 month ago
17 days ago
README

AdaptiveAvgPool3d

English | 简体中文

Applicable Products

Product Support
Ascend 950PR&950DT products √
Atlas A3 products √
Atlas A2 products √
Atlas 200I/500 A2 inference products ×
Atlas inference products √
Atlas training products ×
Kirin X90 processor series products √
Kirin 9030 processor series products √

Function

  • Operator function: When the specified 3D output shape information is given, perform 3D adaptive average pooling on the input tensor.

Parameters

Parameter name Input/Output/Attribute Description Data type Data format
x Input Input parameter for the adaptiveavgpool3d computation. It indicates the target tensor to be converted. FLOAT, FLOAT16, BFLOAT16 ND
output_size Attribute Input parameter for the adaptiveavgpool3d computation. It indicates the shape size of the output in the DHW dimensions. INT64 -
y Output Output parameter of the adaptiveavgpool3d computation. It indicates the converted result tensor. FLOAT, FLOAT16, BFLOAT16 ND
  • Atlas inference products: BFLOAT16 is not supported.
  • Kirin X90/Kirin 9030 processor series products: BFLOAT16 is not supported.

Constraints

None.

Invocation Description

Invocation method Sample code Description
aclnn API test_aclnn_adaptive_avg_pool3d.cpp Invoke the AdaptiveAvgPool3d operator through the aclnnAdaptiveAvgPool3d API.