def _cs_impl(cls):
# todo: set random seed; set 'cache_size'
kernel_linear = Node({"kernel": FixedP("linear")})
kernel_poly = Node(
{
"kernel": FixedP("poly"),
"degree": IntP(uniform, low=0, high=10),
"coef0": RealP(uniform, low=0.0, high=100),
}
)
kernel_rbf = Node({"kernel": FixedP("rbf")})
kernel_sigmoid = Node({"kernel": FixedP("sigmoid"), "coef0": RealP(uniform, low=0.0, high=100),})
kernel_nonlinear = Node(
{"gamma": RealP(uniform, low=0.00001, high=100)}, children=[kernel_poly, kernel_rbf, kernel_sigmoid],
)
top = Node(
{
"C": RealP(uniform, low=1e-4, high=100),
"shrinking": CatP(choice, items=[True, False]),
"probability": FixedP(False),
"tol": OrdP(choice, items=[0.000001, 0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10, 100, 1000, 10000,],),
"class_weight": CatP(choice, items=[None, "balanced"]),
# 'verbose': [False],
"max_iter": FixedP(1000000),
"decision_function_shape": CatP(choice, items=["ovr", "ovo"]),
},
children=[kernel_linear, kernel_nonlinear],
)
return CS(nodes=[top])
def _cs_impl(cls):
# TODO target and prediction
params = {
'function': CatP(choice, items=cls.names()),
'target': CatP(choice, items=['Y']),
'prediction': CatP(choice, items=['Z'])
}
return TransformerCS(Node(params=params))
def _cs_impl(cls):
# TODO target and prediction
params = {
'function': CatP(choice, items=cls.names()),
'input_field': CatP(choice, items=['S']),
'output_field': CatP(choice, items=['S'])
}
return TransformerCS(Node(params=params))
def _cs_impl(cls):
# TODO target and prediction
params = {
"function": CatP(choice, items=cls.names()),
"target": CatP(choice, items=["Y"]),
"prediction": CatP(choice, items=["Z"]),
}
return CS(nodes=[Node(params=params)])
def _cs_impl(cls):
params = {
"criterion": CatP(choice, items=["gini", "entropy"]),
"splitter": FixedP("best"),
"class_weight": CatP(choice, items=[None, "balanced"]),
"max_features": CatP(choice, items=["auto", "sqrt", "log2", None]),
"max_depth": IntP(uniform, low=2, high=1000),
"min_samples_split": RealP(uniform, low=1e-6, high=0.3),
"min_samples_leaf": RealP(uniform, low=1e-6, high=0.3),
"min_weight_fraction_leaf": RealP(uniform, low=0.0, high=0.3),
"min_impurity_decrease": RealP(uniform, low=0.0, high=0.2),
}
return CS(nodes=[Node(params=params)])
def _cs_impl(cls):
params = {
'criterion': CatP(choice, items=['gini', 'entropy']),
'splitter': FixedP('best'),
'class_weight': CatP(choice, items=[None, 'balanced']),
'max_features': CatP(choice, items=['auto', 'sqrt', 'log2', None]),
'max_depth': IntP(uniform, low=2, high=1000),
'min_samples_split': RealP(uniform, low=1e-6, high=0.3),
'min_samples_leaf': RealP(uniform, low=1e-6, high=0.3),
'min_weight_fraction_leaf': RealP(uniform, low=0.0, high=0.3),
'min_impurity_decrease': RealP(uniform, low=0.0, high=0.2)
}
return TransformerCS(nodes=[Node(params=params)])
def _cs_impl(cls):
n_estimators = [100, 500, 1000, 3000, 5000]
params = {
'bootstrap': CatP(choice, items=[True, False]),
'criterion': CatP(choice, items=['gini', 'entropy']),
'max_features': CatP(choice, items=['auto', 'sqrt', 'log2', None]),
'min_impurity_decrease': RealP(uniform, low=0.0, high=0.2),
'min_samples_split': RealP(uniform, low=1e-6, high=0.3),
'min_samples_leaf': RealP(uniform, low=1e-6, high=0.3),
'min_weight_fraction_leaf': RealP(uniform, low=0.0, high=0.3),
'max_depth': IntP(uniform, low=2, high=1000),
'n_estimators': CatP(choice, items=n_estimators),
}
return TransformerCS(nodes=[Node(params=params)])
def _cs_impl(cls):
# todo: set random seed; set 'cache_size'
kernel_linear = Node({'kernel': FixedP('linear')})
kernel_poly = Node({
'kernel': FixedP('poly'),
'degree': IntP(uniform, low=0, high=10),
'coef0': RealP(uniform, low=0.0, high=100)
})
kernel_rbf = Node({'kernel': FixedP('rbf')})
kernel_sigmoid = Node({
'kernel': FixedP('sigmoid'),
'coef0': RealP(uniform, low=0.0, high=100),
})
kernel_nonlinear = Node(
{'gamma': RealP(uniform, low=0.00001, high=100)},
children=[kernel_poly, kernel_rbf, kernel_sigmoid])
top = Node(
{
'C':
RealP(uniform, low=1e-4, high=100),
'shrinking':
CatP(choice, items=[True, False]),
'probability':
FixedP(False),
'tol':
OrdP(choice,
items=[
0.000001, 0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10,
100, 1000, 10000
]),
'class_weight':
CatP(choice, items=[None, 'balanced']),
# 'verbose': [False],
'max_iter':
FixedP(1000000),
'decision_function_shape':
CatP(choice, items=['ovr', 'ovo'])
},
children=[kernel_linear, kernel_nonlinear])
return TransformerCS(nodes=[top])
def _cs_impl(cls):
params = {
'score_func': CatP(
choice,
items=["chi2", "f_classif", "mutual_info_classif"]
),
'k_perc': RealP(uniform, low=0.0, high=1.0)
}
return TransformerCS(nodes=[Node(params=params)])
def _cs_impl(cls):
# TODO target and prediction
params = {
'input_field1': CatP(choice, items=['S']),
'input_field2': CatP(choice, items=['S']),
'direction': CatP(choice, items=[0, 1]),
'output_field': CatP(choice, items=['S'])
}
return TransformerCS(Node(params=params))
# TODO: create a proper test?
# p = Pipeline(
# File('iris.arff'),
# ApplyUsing(NB()),
# Report('$X $Y $Z'),
# MConcat(fields=['X','Y','Z'], output_field='A'),
# Report('$A')
# )
# p.apply()
def _cs_impl(cls):
params = {'feature_range': CatP(choice, items=[(-1, 1), (0, 1)])}
return TransformerCS(nodes=[Node(params=params)])
def _cs_impl(cls, data=None):
params = {
'sampling_strategy':
CatP(choice, items=['not minority', 'not majority', 'all'])
}
return TransformerCS(Node(params=params))
def _cs_impl(cls):
params = {
'engine': CatP(choice, items=['dump', 'mysql', 'sqlite']),
'settings': FixedP({})
}
return TransformerCS(Node(params=params))
def _cs_impl(cls):
params = {
'function': CatP(choice, items=cls.function_from_name.keys()),
'field': CatP(choice, items=['z', 'r', 's'])
}
return TransformerCS(Node(params))
def _cs_impl(cls, data=None):
params = {
'operation': CatP(choice, items=['full', 'translate', 'scale'])
}
return TransformerCS(nodes=[Node(params=params)])
def _cs_impl(cls):
params = {
'distribution': CatP(choice, items=['gaussian', 'bernoulli'])
}
return TransformerCS(nodes=[Node(params=params)])