def objective(trial):
trial.suggest_uniform("DROPOUT", 0.0, 0.5)
trial.suggest_int("EMBEDDING_DIM", 20, 50)
trial.suggest_int("MAX_FILTER_SIZE", 3, 6)
trial.suggest_int("NUM_FILTERS", 16, 32)
trial.suggest_int("HIDDEN_SIZE", 16, 32)
serialization_dir = os.path.join(MODEL_DIR, "test_{}".format(trial.number))
executor = AllenNLPExecutor(trial, CONFIG_PATH, serialization_dir)
return executor.run()
def objective(trial):
trial.suggest_uniform("DROPOUT", 0.0, 0.5)
trial.suggest_int("EMBEDDING_DIM", 20, 50)
trial.suggest_int("MAX_FILTER_SIZE", 3, 6)
trial.suggest_int("NUM_FILTERS", 16, 32)
trial.suggest_int("HIDDEN_SIZE", 16, 32)
config_path = os.path.join(EXAMPLE_DIR, "classifier.jsonnet")
serialization_dir = os.path.join(MODEL_DIR, "test_{}".format(trial.number))
executor = AllenNLPExecutor(trial, config_path, serialization_dir)
return executor.run()
def objective(trial: Trial):
trial.suggest_int("char_embedding_dim", 16, 128)
trial.suggest_int("lstm_hidden_size", 64, 256)
trial.suggest_float("lr", 5e-3, 5e-1, log=True)
executor = AllenNLPExecutor(
trial=trial,
config_file=config_file,
serialization_dir=f"result/{trial.number}",
metrics="best_validation_f1-measure-overall",
include_package="allennlp_models",
)
return executor.run()
def _objective(
trial: Trial,
hparam_path: str,
) -> float:
for hparam in json.load(open(hparam_path)):
attr_type = hparam["type"]
suggest = getattr(trial, "suggest_{}".format(attr_type))
suggest(**hparam["attributes"])
optuna_serialization_dir = os.path.join(
serialization_dir, "trial_{}".format(trial.number))
executor = AllenNLPExecutor(
trial,
config_file,
optuna_serialization_dir,
metrics=metrics,
include_package=include_package,
)
return executor.run()