FilesLast commitLast update
2 months ago
3 months ago
24 days ago
26 days ago
26 days ago
13 days ago
1 month ago
1 month ago
13 days ago
README

BiasAdd

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: adds a bias value to each element in the input tensor.

  • Computation formula:

outi=xi+biasiout_i= x_i + bias_i

Parameter Description

Parameter Name Input/Output/Attribute Description Data Type Data Format
x Input Input parameter for the BiasAdd computation, which is x_i in the formula. FLOAT, FLOAT16, BFLOAT16, INT32, INT64 NCHW, NHWC, NDHWC, NCDHW, ND
bias Input Accumulated bias, which is bias_i in the formula. FLOAT, FLOAT16, BFLOAT16, INT32, INT64 ND
out Output Output parameter for the BiasAdd computation, which is out_i in the formula. FLOAT, FLOAT16, BFLOAT16, INT32, INT64 NCHW, NHWC, NDHWC, NCDHW, ND
  • Atlas training products and Atlas inference products: BFLOAT16 is not supported.

Constraint Description

None

Invocation Description

Invocation Method Invocation Sample Description
Graph-mode invocation test_geir_bias_add Invokes the BiasAdd operator by building a graph through the operator IR.