cuRAND (nvmath.bindings.curand)#

For detailed documentation on the original C APIs, please refer to cuRAND documentation.

Enums and constants#

RngType(value[, names, module, qualname, ...])

See curandRngType.

Ordering(value[, names, module, qualname, ...])

See curandOrdering.

Method(value[, names, module, qualname, ...])

See curandMethod.

Status(value[, names, module, qualname, ...])

See curandStatus.

cuRANDError(status)

Functions#

create_generator(int rng_type)

Create new random number generator.

create_generator_host(int rng_type)

Create new host CPU random number generator.

destroy_generator(intptr_t generator)

Destroy an existing generator.

get_version()

Return the version number of the library.

get_property(int type)

Return the value of the curand property.

set_stream(intptr_t generator, intptr_t stream)

Set the current stream for CURAND kernel launches.

set_pseudo_random_generator_seed(...)

Set the seed value of the pseudo-random number generator.

set_generator_offset(intptr_t generator, ...)

Set the absolute offset of the pseudo or quasirandom number generator.

set_generator_ordering(intptr_t generator, ...)

Set the ordering of results of the pseudo or quasirandom number generator.

set_quasi_random_generator_dimensions(...)

Set the number of dimensions.

generate(intptr_t generator, ...)

Generate 32-bit pseudo or quasirandom numbers.

generate_long_long(intptr_t generator, ...)

Generate 64-bit quasirandom numbers.

generate_uniform(intptr_t generator, ...)

Generate uniformly distributed floats.

generate_uniform_double(intptr_t generator, ...)

Generate uniformly distributed doubles.

generate_normal(intptr_t generator, ...)

Generate normally distributed doubles.

generate_normal_double(intptr_t generator, ...)

Generate normally distributed doubles.

generate_log_normal(intptr_t generator, ...)

Generate log-normally distributed floats.

generate_log_normal_double(...)

Generate log-normally distributed doubles.

create_poisson_distribution(double lambda_, ...)

Construct the histogram array for a Poisson distribution.

destroy_distribution(...)

Destroy the histogram array for a discrete distribution (e.g. Poisson).

generate_poisson(intptr_t generator, ...)

Generate Poisson-distributed unsigned ints.

generate_poisson_method(intptr_t generator, ...)

generate_binomial(intptr_t generator, ...)

generate_binomial_method(intptr_t generator, ...)

generate_seeds(intptr_t generator)

Setup starting states.