qns.utils package

Submodules

qns.utils.log module

qns.utils.log.critical(msg, *args)

qns.utils.log.debug(msg, *args)

qns.utils.log.error(msg, *args)

qns.utils.log.info(msg, *args)

qns.utils.log.install(s)

Install the logger to the simulator

Parameters:

s (Simulator) – the simulator

qns.utils.log.logger = <Logger qns (INFO)>

The default logger used by SimQN

qns.utils.log.monitor(*args, sep: str = ',', with_time: bool = False)

qns.utils.log.warn(msg, *args)

qns.utils.multiprocess module

class qns.utils.multiprocess.MPSimulations(settings: Dict = {}, iter_count: int = 1, aggregate: bool = True, cores: int = -1, name: str | None = None)

Bases: object

MultiProcessSimulations will help users to perfrom multiple simulations with different experiment settings and leverage multiple processes.

get_data()

Get the simulation results

Returns:

a result data in pd.DataFrame

get_raw_data()

Get the original raw results, no matter aggregate is True or not.

Returns:

a result data in pd.DataFrame

prepare_setting()

Generate the experiment setting for each experiments.

run(setting: Dict = {}) → Dict

This function should be overwited by users to provide codes that run a single simulation.

Parameters:

setting (Dict) – the simulation setting, e.g. {‘node_num’: 10, ‘req_num’: 10, ‘memory_size’: 50}

Returns:

100, ‘fidelity’: 0.88}

Return type:

a dictionary that contains all results, e.g. {‘throughput’

start()

Start the multiple process simulation

qns.utils.rnd module

qns.utils.rnd.get_choice(a)

return an random element from a list

Parameters:

a – a iterable object

qns.utils.rnd.get_normal(mean: float = 0, std: float = 1)

qns.utils.rnd.get_rand(low: float = 0, high: float = 1) → float

Get a random number from [low, high)

Parameters:

• low (int) – the low bound

• high (int) – the high bound

qns.utils.rnd.get_randint(low: int, high: int) → float

Get a random integer from [low, high]

Parameters:

• low (int) – the low bound

• high (int) – the high bound

qns.utils.rnd.get_weighted_choice(a, weights, k=1)

return k random element from a list with weights

Parameters:

• a – a iterable object

• weights – a list of weights

• k – the number of random sample elements

qns.utils.rnd.set_seed(seed: int | None = None)

Set a seed for random generator

Parameters:

seed (int) – the seed

Module contents

qns.utils.critical(msg, *args)

qns.utils.debug(msg, *args)

qns.utils.error(msg, *args)

qns.utils.get_choice(a)

return an random element from a list

Parameters:

a – a iterable object

qns.utils.get_rand(low: float = 0, high: float = 1) → float

Get a random number from [low, high)

Parameters:

• low (int) – the low bound

• high (int) – the high bound

qns.utils.get_randint(low: int, high: int) → float

Get a random integer from [low, high]

Parameters:

• low (int) – the low bound

• high (int) – the high bound

qns.utils.info(msg, *args)

qns.utils.install(s)

Install the logger to the simulator

Parameters:

s (Simulator) – the simulator

qns.utils.monitor(*args, sep: str = ',', with_time: bool = False)

qns.utils.set_seed(seed: int | None = None)

Set a seed for random generator

Parameters:

seed (int) – the seed

qns.utils.warn(msg, *args)