def test_random_oversample():
# float
X_new, y_new = resreg.random_oversample(X, y, relevance=relevance,
relevance_threshold=0.5, over=0.75,
random_state=0)
assert round(np.sum(y_new), 3) == 4.971
# balance
X_new, y_new = resreg.random_oversample(X, y, relevance=relevance,
relevance_threshold=0.5, over='balance',
random_state=0)
assert round(np.sum(y_new), 3) == 5.090
# extreme
X_new, y_new = resreg.random_oversample(X, y, relevance=relevance,
relevance_threshold=0.5, over='extreme',
random_state=0)
assert round(np.sum(y_new), 3) == -49.722
# average
X_new, y_new = resreg.random_oversample(X, y, relevance=relevance,
relevance_threshold=0.5, over='average',
random_state=0)
assert round(np.sum(y_new), 3) == -41.166
# errors
with pytest.raises(Exception):
X_new, y_new = resreg.random_oversample(X, y, relevance=relevance,
relevance_threshold=0.5, over=2)
bin_test_size=70,
verbose=False,
random_state=rrr)
X_train, y_train = X[train_indices, :], y[train_indices]
X_test, y_test = X[test_indices, :], y[test_indices]
# Unpack hyperparameters, resample training data, and fit regressors
reg = DecisionTreeRegressor(random_state=rrr) if 'REBAGG' in strategy else \
RandomForestRegressor(n_estimators=10, n_jobs=-1, random_state=rrr)
if strategy == 'RO':
cl, ch, sample_method = param
relevance = resreg.sigmoid_relevance(y_train, cl=cl, ch=ch)
X_train, y_train = resreg.random_oversample(X_train,
y_train,
relevance,
relevance_threshold=0.5,
over=sample_method,
random_state=rrr)
reg.fit(X_train, y_train)
elif strategy == 'SMOTER':
cl, ch, sample_method, k = param
relevance = resreg.sigmoid_relevance(y_train, cl=cl, ch=ch)
X_train, y_train = resreg.smoter(X_train,
y_train,
relevance,
relevance_threshold=0.5,
k=k,
over=sample_method,
random_state=rrr)
reg.fit(X_train, y_train)
random_state=0)
X_train, y_train = X[train_indices, :], y[train_indices]
if strategy == 'None':
sns.kdeplot(y_train,
bw=7,
linewidth=lw,
linestyle=style1,
color='black',
label='TOME')
elif strategy == 'RO':
relevance = resreg.sigmoid_relevance(y_train, cl=None, ch=60)
X_train, y_train = resreg.random_oversample(X_train,
y_train,
relevance=relevance,
relevance_threshold=0.5,
over='balance',
random_state=0)
sns.kdeplot(y_train,
bw=7,
linewidth=lw,
label=strategy,
color='blue',
linestyle=style2)
elif strategy == 'SMOTER':
relevance = resreg.sigmoid_relevance(y_train, cl=None, ch=60)
X_train, y_train = resreg.smoter(X_train,
y_train,
relevance=relevance,
relevance_threshold=0.5,
