dequantized(to:scale:bias:axis:)
Converts a tensor you quantize to a 32-bit floating-point tensor.
Declaration
func dequantized(to type: MLCDataType, scale: MLCTensor, bias: MLCTensor, axis: Int) -> MLCTensor?Parameters
- type:
The tensor data type.
- scale:
The scale the system uses when quantizing the data.
- bias:
The offset value the system uses when quantizing the data.
- axis:
The dimension on which to apply per-channel quantization.
Return Value
A tensor the system dequantizes.