def _execute():
env = Environment(
train_dataset=get_toy_classification_data(target='diagnosis'),
root_results_path='HyperparameterHunterAssets',
target_column='diagnosis',
metrics_map=['roc_auc_score'],
cross_validation_type=RepeatedStratifiedKFold,
cross_validation_params=dict(n_splits=5, n_repeats=2, random_state=32),
)
optimizer = ExtraTreesOptimization(
iterations=10,
read_experiments=True,
random_state=None,
)
optimizer.set_experiment_guidelines(
model_initializer=RGFClassifier,
model_init_params=dict(max_leaf=1000,
algorithm=Categorical(
['RGF', 'RGF_Opt', 'RGF_Sib']),
l2=Real(0.01, 0.3),
normalize=Categorical([True, False]),
learning_rate=Real(0.3, 0.7),
loss=Categorical(['LS', 'Expo', 'Log', 'Abs'])),
)
optimizer.go()
def opt_dtc_0():
optimizer = ExtraTreesOptimization(iterations=2, random_state=1337)
optimizer.set_experiment_guidelines(
model_initializer=DecisionTreeClassifier,
model_init_params=dict(
criterion="gini",
min_samples_split=Integer(2, 5),
splitter=Categorical(["best", "random"]),
min_weight_fraction_leaf=Real(0.0, 0.1),
),
)
optimizer.go()
yield optimizer
target_column="median_value",
metrics=dict(r2=r2_score),
cv_type=RepeatedKFold,
cv_params=dict(n_repeats=2, n_splits=5, random_state=42),
)
# Now that HyperparameterHunter has an active `Environment`, we can do two things:
#################### 1. Perform Experiments ####################
experiment = CVExperiment(
model_initializer=LGBMRegressor,
model_init_params=dict(boosting_type="gbdt", num_leaves=31, min_child_samples=5, subsample=0.5),
)
# And/or...
#################### 2. Hyperparameter Optimization ####################
optimizer = ExtraTreesOptimization(iterations=12, random_state=1337)
optimizer.set_experiment_guidelines(
model_initializer=LGBMRegressor,
model_init_params=dict(
boosting_type=Categorical(["gbdt", "dart"]),
num_leaves=Integer(10, 40),
max_depth=-1,
min_child_samples=5,
subsample=Real(0.3, 0.7),
),
)
optimizer.go()
# Notice, `optimizer` recognizes our earlier `experiment`'s hyperparameters fit inside the search
# space/guidelines set for `optimizer`.
cross_validation_type="KFold",
cross_validation_params=dict(n_splits=10, random_state=42),
runs=3,
)
# Now that HyperparameterHunter has an active `Environment`, we can do two things:
#################### 1. Perform Experiments ####################
experiment = CVExperiment(
model_initializer=RGFRegressor,
model_init_params=dict(max_leaf=2000, algorithm="RGF",
min_samples_leaf=10),
)
# And/or...
#################### 2. Hyperparameter Optimization ####################
optimizer = ExtraTreesOptimization(iterations=30, random_state=42)
optimizer.set_experiment_guidelines(
model_initializer=RGFRegressor,
model_init_params=dict(
max_leaf=2000,
algorithm=Categorical(["RGF", "RGF_Opt", "RGF_Sib"]),
l2=Real(0.01, 0.3),
normalize=Categorical([True, False]),
learning_rate=Real(0.3, 0.7),
loss=Categorical(["LS", "Expo", "Log"]),
),
)
optimizer.go()
# Notice, `optimizer` recognizes our earlier `experiment`'s hyperparameters fit inside the search
# space/guidelines set for `optimizer`.