def load_data():
set_random_state(9527)
df = dsutils.load_bank().head(3000)
encoder = MultiLabelEncoder()
df = encoder.fit_transform(df)
df.drop(['id'], axis=1, inplace=True)
return df
def test_set_random_state(self):
from hypernets.core import set_random_state
set_random_state(9527)
searcher = RandomSearcher(self.get_space)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[98, 0, 0, 0.96], [9, 0, 0, 0.93], [60, 0, 1, 0.24],
[54, 0, 1, 0.7], [25, 0, 1, 0.73], [67, 1, 1, 0.43],
[57, 1, 1, 0.05], [49, 0, 0, 0.71],
[71, 1, 1, 0.49]]
set_random_state(None)
searcher = RandomSearcher(self.get_space)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors != [[98, 0, 0, 0.96], [9, 0, 0, 0.93], [60, 0, 1, 0.24],
[54, 0, 1, 0.7], [25, 0, 1, 0.73], [67, 1, 1, 0.43],
[57, 1, 1, 0.05], [49, 0, 0, 0.71],
[71, 1, 1, 0.49]]
set_random_state(9527)
searcher = RandomSearcher(self.get_space)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[98, 0, 0, 0.96], [9, 0, 0, 0.93], [60, 0, 1, 0.24],
[54, 0, 1, 0.7], [25, 0, 1, 0.73], [67, 1, 1, 0.43],
[57, 1, 1, 0.05], [49, 0, 0, 0.71],
[71, 1, 1, 0.49]]
set_random_state(1)
searcher = RandomSearcher(self.get_space)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[38, 1, 0, 0.93], [10, 1, 1, 0.15],
[17, 1, 0, 0.39], [7, 1, 0, 0.85], [19, 0, 1, 0.44],
[29, 1, 0, 0.67], [88, 1, 1, 0.43], [95, 0, 0, 0.8],
[10, 1, 1, 0.09]]
set_random_state(None)
def test_set_random_state(self):
from hypernets.core import set_random_state
set_random_state(9527)
searcher = MCTSSearcher(self.get_space, max_node_space=10)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[98, 0, 1, 0.86], [2, 0, 1, 0.58], [2, 0, 1, 0.73],
[2, 0, 1, 0.58], [2, 1, 0, 0.67], [2, 0, 1, 0.01],
[2, 0, 1, 0.22], [2, 0, 1, 0.55], [2, 0, 1, 0.32]]
set_random_state(None)
searcher = MCTSSearcher(self.get_space, max_node_space=10)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors != [[98, 0, 1, 0.86], [2, 0, 1, 0.58], [2, 0, 1, 0.73],
[2, 0, 1, 0.58], [2, 1, 0, 0.67], [2, 0, 1, 0.01],
[2, 0, 1, 0.22], [2, 0, 1, 0.55], [2, 0, 1, 0.32]]
set_random_state(9527)
searcher = MCTSSearcher(self.get_space, max_node_space=10)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[98, 0, 1, 0.86], [2, 0, 1, 0.58], [2, 0, 1, 0.73],
[2, 0, 1, 0.58], [2, 1, 0, 0.67], [2, 0, 1, 0.01],
[2, 0, 1, 0.22], [2, 0, 1, 0.55], [2, 0, 1, 0.32]]
set_random_state(1)
searcher = MCTSSearcher(self.get_space, max_node_space=10)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[73, 1, 0, 0.4], [2, 0, 0, 0.42], [2, 0, 0, 0.2],
[2, 0, 0, 0.03], [2, 0, 0, 0.42], [2, 1, 1, 0.14],
[2, 0, 0, 0.8], [2, 1, 1, 0.31], [2, 0, 1, 0.88]]
set_random_state(None)
def test_basic(self):
df = self.load_data()
y = df.pop('y')
tb = get_tool_box(df)
X_train, X_test, y_train, y_test = tb.train_test_split(df,
y,
test_size=0.3,
random_state=42)
hm, _ = train(X_train, y_train, X_test, y_test, max_trials=5)
best_trials = hm.get_top_trials(3)
estimators = [
hm.load_estimator(trial.model_file) for trial in best_trials
]
importances = tb.permutation_importance_batch(
estimators,
X_test,
y_test,
get_scorer('roc_auc_ovr'),
n_jobs=1,
n_repeats=5,
random_state=get_random_state())
feature_index = np.argwhere(importances.importances_mean < 1e-5)
selected_features = [
feat for i, feat in enumerate(X_train.columns.to_list())
if i not in feature_index
]
unselected_features = [
c for c in X_train.columns.to_list() if c not in selected_features
]
set_random_state(None)
print('selected: ', selected_features)
print('unselected:', unselected_features)
def test_set_random_state(self):
from hypernets.core import set_random_state
set_random_state(9527)
searcher = EvolutionSearcher(
get_space,
5,
3,
regularized=False,
optimize_direction=OptimizeDirection.Maximize)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[98, 0, 0, 0.96], [9, 0, 0, 0.93], [60, 0, 1, 0.24],
[54, 0, 1, 0.7], [25, 0, 1, 0.73], [67, 1, 1, 0.43],
[57, 1, 1, 0.05], [49, 0, 0, 0.71],
[71, 1, 1, 0.49]]
set_random_state(None)
searcher = EvolutionSearcher(
get_space,
5,
3,
regularized=False,
optimize_direction=OptimizeDirection.Maximize)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors != [[98, 0, 0, 0.96], [9, 0, 0, 0.93], [60, 0, 1, 0.24],
[54, 0, 1, 0.7], [25, 0, 1, 0.73], [67, 1, 1, 0.43],
[57, 1, 1, 0.05], [49, 0, 0, 0.71],
[71, 1, 1, 0.49]]
set_random_state(9527)
searcher = EvolutionSearcher(
get_space,
5,
3,
regularized=False,
optimize_direction=OptimizeDirection.Maximize)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[98, 0, 0, 0.96], [9, 0, 0, 0.93], [60, 0, 1, 0.24],
[54, 0, 1, 0.7], [25, 0, 1, 0.73], [67, 1, 1, 0.43],
[57, 1, 1, 0.05], [49, 0, 0, 0.71],
[71, 1, 1, 0.49]]
set_random_state(1)
searcher = EvolutionSearcher(
get_space,
5,
3,
regularized=False,
optimize_direction=OptimizeDirection.Maximize)
vectors = []
for i in range(1, 10):
vectors.append(searcher.sample().vectors)
assert vectors == [[38, 1, 0, 0.93], [10, 1, 1, 0.15],
[17, 1, 0, 0.39], [7, 1, 0, 0.85], [19, 0, 1, 0.44],
[29, 1, 0, 0.67], [88, 1, 1, 0.43], [95, 0, 0, 0.8],
[10, 1, 1, 0.09]]
set_random_state(None)
def train_heart_disease(**kwargs):
from hypernets.tabular.datasets import dsutils
from sklearn.model_selection import train_test_split
X = dsutils.load_heart_disease_uci()
y = X.pop('target')
X_train, X_test, y_train, y_test = \
train_test_split(X, y, test_size=0.3, random_state=randint())
X_train, X_eval, y_train, y_eval = \
train_test_split(X_train, y_train, test_size=0.3, random_state=randint())
kwargs = {'reward_metric': 'auc', 'max_trials': 10, **kwargs}
hm, model = train(X_train, y_train, X_eval, y_eval, const.TASK_BINARY, **kwargs)
print('-' * 50)
scores = model.evaluate(X_test, y_test, metrics=['auc', 'accuracy', 'f1', 'recall', 'precision'])
print('scores:', scores)
trials = hm.get_top_trials(10)
models = [hm.load_estimator(t.model_file) for t in trials]
msgs = [f'{t.trial_no},{t.reward},{m.cls.__name__} {m.model_args}' for t, m in zip(trials, models)]
print('top trials:')
print('\n'.join(msgs))
if __name__ == '__main__':
set_random_state(335)
train_heart_disease()