libcity.executor.hyper_tuning¶

exception libcity.executor.hyper_tuning.ExhaustiveSearchError[source]¶: Bases: Exception

class libcity.executor.hyper_tuning.HyperTuning(objective_function, space=None, params_file=None, algo='grid_search', max_evals=100, task=None, model_name=None, dataset_name=None, config_file=None, saved_model=True, train=True, other_args=None)[source]¶

Bases: object

自动调参

Note

HyperTuning is based on the hyperopt (https://github.com/hyperopt/hyperopt)

https://github.com/hyperopt/hyperopt/issues/200

fn(params)[source]¶

static params2str(params)[source]¶

save_result(filename=None)[source]¶

start()[source]¶

libcity.executor.hyper_tuning.exhaustive_search(new_ids, domain, trials, seed, nb_max_sucessive_failures=1000)[source]¶