Convert type🔗

Changes the data type of the input object or converts a string to a number.

Input🔗

input: A scalar, a matrix, a tensor or a string. Accepted types depend on outputType: a scalar output type requires a scalar or string as an input, a matrix output type requires matrix input etc.
outputType: The type of the output.
roundingMode: The way floating-point values are converted to integers.

Outputs🔗

output: The converted value.

enum OutputType🔗

Type of the output is scalar, matrix or tensor.

Values:

enumerator Bool🔗

enumerator Int32🔗

enumerator Float🔗

enumerator Double🔗

enumerator FComplex🔗

enumerator DComplex🔗

enumerator BoolMatrix🔗

enumerator Int32Matrix🔗

enumerator FloatMatrix🔗

enumerator DoubleMatrix🔗

enumerator FComplexMatrix🔗

enumerator DComplexMatrix🔗

enumerator BoolTensor🔗

enumerator Int8Tensor🔗

enumerator Uint8Tensor🔗

enumerator Int16Tensor🔗

enumerator Uint16Tensor🔗

enumerator Int32Tensor🔗

enumerator Uint32Tensor🔗

enumerator Int64Tensor🔗

enumerator Uint64Tensor🔗

enumerator FloatTensor🔗

enumerator Float16Tensor🔗

enumerator DoubleTensor🔗

enumerator FComplexTensor🔗

enumerator DComplexTensor🔗

enum RoundingMode🔗

Rounding and clamping modes.

Rounding takes an effect only when the the input type is a floating point type and the output type is an integral type. Clamping takes an effect only when the output type is an integral type.

Values:

enumerator TruncateAndWrap🔗: Truncate the decimal part and cast the result to the output type without overflow or underflow checks.

enumerator RoundAndWrap🔗: Round the decimal part and and cast the result to the output type without overflow or underflow checks.

enumerator TruncateAndClamp🔗: Truncate the decimal part and clamp the result between minimum and maximum values of the output type to prevent spurious overflow and underflow.

enumerator RoundAndClamp🔗: Round the decimal part and clamp the result between minimum and maximum values of the output type to prevent spurious overflow and underflow.