jax.numpy.finfo

jax.numpy.finfo(dtype)[source]

Machine limits for floating point types.

LAX-backend implementation of finfo(). Original docstring below.

finfo(dtype)

jax.numpy.bits

The number of bits occupied by the type.

Type

int

jax.numpy.eps

The difference between 1.0 and the next smallest representable float larger than 1.0. For example, for 64-bit binary floats in the IEEE-754 standard, eps = 2**-52, approximately 2.22e-16.

Type

float

jax.numpy.epsneg

The difference between 1.0 and the next smallest representable float less than 1.0. For example, for 64-bit binary floats in the IEEE-754 standard, epsneg = 2**-53, approximately 1.11e-16.

Type

float

jax.numpy.iexp

The number of bits in the exponent portion of the floating point representation.

Type

int

jax.numpy.machar

The object which calculated these parameters and holds more detailed information.

Type

MachAr

jax.numpy.machep

The exponent that yields eps.

Type

int

jax.numpy.max[source]

The largest representable number.

Type

floating point number of the appropriate type

jax.numpy.maxexp

The smallest positive power of the base (2) that causes overflow.

Type

int

jax.numpy.min[source]

The smallest representable number, typically -max.

Type

floating point number of the appropriate type

jax.numpy.minexp

The most negative power of the base (2) consistent with there being no leading 0’s in the mantissa.

Type

int

jax.numpy.negep

The exponent that yields epsneg.

Type

int

jax.numpy.nexp

The number of bits in the exponent including its sign and bias.

Type

int

jax.numpy.nmant

The number of bits in the mantissa.

Type

int

jax.numpy.precision

The approximate number of decimal digits to which this kind of float is precise.

Type

int

jax.numpy.resolution

The approximate decimal resolution of this type, i.e., 10**-precision.

Type

floating point number of the appropriate type

jax.numpy.tiny

The smallest positive usable number. Type of tiny is an appropriate floating point type.

Type

float

Parameters

dtype (float, dtype, or instance) – Kind of floating point data-type about which to get information..