Utility functions¶

Create scalars and arrays¶

apytypes.fx(value: int | float, int_bits: int | None = None, frac_bits: int | None = None, bits: int | None = None, *, force_complex: Literal[False] = False) → APyFixed¶

apytypes.fx(value: int | float, int_bits: int | None = None, frac_bits: int | None = None, bits: int | None = None, *, force_complex: Literal[True]) → APyCFixed

apytypes.fx(value: complex, int_bits: int | None = None, frac_bits: int | None = None, bits: int | None = None, *, force_complex: bool = False) → APyCFixed

apytypes.fx(value: Iterable[Any], int_bits: int | None = None, frac_bits: int | None = None, bits: int | None = None, *, force_complex: Literal[False] = False) → APyFixedArray

apytypes.fx(value: Iterable[Any], int_bits: int | None = None, frac_bits: int | None = None, bits: int | None = None, *, force_complex: Literal[True]) → APyCFixedArray

Create a fixed-point scalar or array.

Convenience function that passes value and bit-specifiers to one of:

depending on value. Returns APyFixed when value is int or float. Returns APyCFixed when value is complex or when force_complex is True. Returns APyFixedArray or APyCFixedArray when value is a sequence of numbers.

Added in version 0.3.

Parameters:

valint, float, list of int or float, ndarray: Floating-point/integer/complex value(s) to initialize from.
int_bitsint, optional: Number of integer bits in the created fixed-point object.
frac_bitsint, optional: Number of fractional bits in the created fixed-point object.
bitsint, optional: Total number of bits in the created fixed-point object.
force_complexbool, default: False: If True, force the return value to be APyCFixed or APyCFixedArray, even if value is real.

Returns:

APyFixed, APyFixedArray, APyCFixed, or
APyCFixedArray

apytypes.fp(value: int | float, exp_bits: int, man_bits: int, bias: int | None = None, *, force_complex: Literal[False] = False) → APyFloat¶

apytypes.fp(value: int | float, exp_bits: int, man_bits: int, bias: int | None = None, *, force_complex: Literal[True]) → APyCFloat

apytypes.fp(value: complex, exp_bits: int, man_bits: int, bias: int | None = None, *, force_complex: bool = False) → APyCFloat

apytypes.fp(value: Iterable[Any], exp_bits: int, man_bits: int, bias: int | None = None, *, force_complex: Literal[False] = False) → APyFloatArray

apytypes.fp(value: Iterable[Any], exp_bits: int, man_bits: int, bias: int | None = None, *, force_complex: Literal[True]) → APyCFloatArray

Create a floating-point scalar or array.

Convenience function that passes value and bit-specifiers to one of:

depending on value. Returns APyFloat when value is int or float. Returns APyCFloat when value is complex or when force_complex is True. Returns APyFloatArray or APyCFloatArray when value is a sequence of numbers.

Added in version 0.3.

Parameters:

valueint, float, list of int or float, ndarray: Floating-point/integer/complex value(s) to initialize from.
exp_bitsint: Number of exponent bits.
man_bitsint: Number of mantissa bits.
biasint, optional: Exponent bias. If not provided, bias is 2**(exp_bits - 1) - 1.
force_complexbool, default: False: If True, force the return value to be APyCFloat or APyCFloatArray, even if value is real.

Returns:

APyFloat, APyFloatArray, APyCFloat, or
APyCFloatArray

apytypes.from_bits(bit_patterns: int, *, int_bits: int | None = None, frac_bits: int | None = None, bits: int | None = None, exp_bits: None = None, man_bits: None = None, bias: None = None, force_complex: Literal[False] = False) → APyFixed¶

apytypes.from_bits(bit_patterns: Iterable[Any], *, int_bits: int | None = None, frac_bits: int | None = None, bits: int | None = None, exp_bits: None = None, man_bits: None = None, bias: None = None, force_complex: Literal[False] = False) → APyFixedArray

apytypes.from_bits(bit_patterns: int, *, int_bits: None = None, frac_bits: None = None, bits: None = None, exp_bits: int, man_bits: int, bias: int | None = None, force_complex: Literal[False] = False) → APyFloat

apytypes.from_bits(bit_patterns: Iterable[Any], *, int_bits: None = None, frac_bits: None = None, bits: None = None, exp_bits: int, man_bits: int, bias: int | None = None, force_complex: Literal[False] = False) → APyFloatArray

Create an APyType scalar or array from bit pattern.

Added in version 0.6.

Parameters:

bit_patternsint or Iterable: Bit representation of the number(s). Number of rows and columns in output.
int_bitsint, optional: Number of fixed-point integer bits.
frac_bitsint, optional: Number of fixed-point fractional bits.
bitsint, optional: Number of fixed-point bits.
exp_bitsint, optional: Number of floating-point exponential bits.
man_bitsint, optional: Number of floating-point mantissa bits.
biasint, optional: Exponent bias. If not provided, bias is 2**(exp_bits - 1) - 1.
force_complexbool, optional: Returns a complex-valued type.

Returns:

resultAPyFixed, APyFloat, APyFixedArray, APyFloatArray, APyCFixed, APyCFloat, APyCFixedArray or APyCFloatArray

Evaluate functions¶

Utility function to evaluate functions on arguments and convert back.

This does exactly:

Convert argument(s) to float
Evaluate function with float argument(s)
Convert result back to the same type as the argument

Hence, there may be numerical issues, but it provides a simple way to perform, e.g., sin, assuming that it is based on a look-up table.

Added in version 0.3.

Parameters:

fncallable: The function to evaluate.
argsAPyFixed, APyFloat, APyFixedArray, APyFloatArray: The argument(s) to evaluate the function for.

Returns:

APyFloat, APyFloatArray, APyFixed, APyFixedArray

Examples

>>> import apytypes as apy
>>> import math
>>> a = apy.APyFixed(19, 1, 5)
>>> apy.fn(math.sin, a)
APyFixed(18, int_bits=1, frac_bits=5)

Enum types¶

class apytypes.ConvolutionMode(*values)¶

FULL = 0¶

SAME = 1¶

VALID = 2¶