FilesLast commitLast update
17 days ago
8 months ago
1 month ago
10 days ago
2 days ago
30 days ago
10 days ago
2 days ago
28 days ago
9 days ago
README

MaskedScale

English | 简体中文

Product Support Status

Product Supported
Ascend 950PR&950DT products √
Atlas A3 products ×
Atlas A2 products ×
Atlas 200I/500 A2 inference products ×
Atlas inference products ×
Atlas training products ×

Function Description

  • Operator function: performs elementwise masked scale computation.

  • Computation formula:

    y=x×mask×scaley = x \times mask \times scale

Parameter Description

Parameter Name Input/Output/Attribute Description Data Type Data Format
x Input Input tensor, which is x in the formula. FLOAT16, BFLOAT16, FLOAT ND
mask Input Mask tensor, which is mask in the formula. The shape is the same as that of x. INT8, UINT8, FLOAT16, FLOAT ND
scale Attribute Scale factor, which is scale in the formula. FLOAT -
y Output Output tensor, which is y in the formula. The data type and shape are the same as those of x. FLOAT16, BFLOAT16, FLOAT ND

Constraint Description

None

Invocation Description

Invocation Method Invocation Sample Description
aclnn invocation test_aclnn_masked_scale Invokes the MaskedScale operator through the aclnnMaskedScale API.