Abs
Description
Returns the absolute value of the input tensor. The absolute value y = abs(x) is applied to the tensor elements.
Input
x: input tensor of type float, float16, int8, int16, int32, or int64. The data format is ND.
Output
y: output tensor. Has the identical data type and shape as the input. The data format is ND.
Constraints
If the computation workload is too large, it may cause the operator execution to time out (an aicore error is reported, with errorStr being: timeout or trap error). This scenario occurs when the last two axes are combined into fewer than 16 axes, and the previous axes are combined into more than 16 axes.
ONNX Opset Support
Opset v8/v9/v10/v11/v12/v13/v14/v15/v16/v17/v18
Parent topic: Supported ONNX Operators