def rfecv_sklearn_example(image="rfecv_sklearn_example.png"):
X, y = make_classification(
n_samples=1000, n_features=25, n_informative=3, n_redundant=2,
n_repeated=0, n_classes=8, n_clusters_per_class=1, random_state=0
)
_, ax = plt.subplots()
oz = RFECV(SVC(kernel='linear', C=1), ax=ax)
oz.fit(X, y)
oz.poof(outpath=os.path.join(IMAGES, image))
def rfecv_credit_example(image="rfecv_credit.png"):
data = pd.read_csv(os.path.join(DATA, "credit", "credit.csv"))
target = "default"
features = [col for col in data.columns if col != target]
X = data[features]
y = data[target]
_, ax = plt.subplots()
cv = StratifiedKFold(5)
oz = RFECV(RandomForestClassifier(), ax=ax, cv=cv, scoring='f1_weighted')
oz.fit(X, y)
oz.poof(outpath=os.path.join(IMAGES, image))
def fazer_selecao_features_rfe(modelo):
features = dados_completo.columns
rfe = RFECV(modelo, cv=kfold, scoring=scoring)
rfe.fit(dados_completo.drop(['classe'], axis=1),
dados_completo['classe'].values)
print(rfe.poof())
print("Caraterísticas ordenadas pelo rank RFE:")
print(sorted(zip(map(lambda x: round(x, 4), rfe.ranking_), features)))
ranking = sorted(zip(rfe.support_, features))
print("Características selecionadas", ranking)
return rfe.transform(dados_completo.drop(['classe'], axis=1))
def rfecv_credit_example(image="rfecv_credit.png"):
data = pd.read_csv(os.path.join(DATA, "credit", "credit.csv"))
target = "default"
features = [col for col in data.columns if col != target]
X = data[features]
y = data[target]
_, ax = plt.subplots()
cv = StratifiedKFold(5)
oz = RFECV(RandomForestClassifier(), ax=ax, cv=cv, scoring='f1_weighted')
oz.fit(X, y)
oz.poof(outpath=os.path.join(IMAGES, image))
def fazer_selecao_features_rfe():
features = X_treino.columns
rfe = RFECV(RandomForestClassifier(random_state=random_state,
oob_score=True),
cv=kfold,
scoring='accuracy')
rfe.fit(X_treino, Y_treino.values.ravel())
print(rfe.poof())
print("Caraterísticas ordenadas pelo rank RFE:")
print(sorted(zip(map(lambda x: round(x, 4), rfe.ranking_), features)))
ranking = sorted(zip(rfe.support_, features))
print("Características selecionadas", ranking)
return rfe.transform(X_treino)
def rfecv_sklearn_example(image="rfecv_sklearn_example.png"):
X, y = make_classification(n_samples=1000,
n_features=25,
n_informative=3,
n_redundant=2,
n_repeated=0,
n_classes=8,
n_clusters_per_class=1,
random_state=0)
_, ax = plt.subplots()
oz = RFECV(SVC(kernel='linear', C=1), ax=ax)
oz.fit(X, y)
oz.poof(outpath=os.path.join(IMAGES, image))
def feature_selection(model, features, X, y):
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import StratifiedKFold
from yellowbrick.features import RFECV
# df = load_data('credit')
# features = [col for col in data.columns if col != target]
# X = data[features]
# y = data[target]
cv = StratifiedKFold(5)
oz = RFECV(model, cv=cv, scoring='f1_weighted')
oz.fit(X, y)
oz.poof()
def fazer_selecao_features_rfe():
features = X_treino.columns
rfe = RFECV(RandomForestClassifier(random_state=random_state,
oob_score=True,
n_estimators=250,
criterion='entropy',
max_depth=75,
max_features='log2',
min_samples_leaf=1,
min_samples_split=2),
cv=kfold,
scoring='accuracy')
rfe.fit(X_treino, Y_treino.values.ravel())
print(rfe.poof())
print("Caraterísticas ordenadas pelo rank RFE:")
print(sorted(zip(map(lambda x: round(x, 4), rfe.ranking_), features)))
ranking = sorted(zip(rfe.support_, features))
print("Características selecionadas", ranking)
return rfe.transform(X_treino)
# %%
feat = feature_names[feat][np.mean(abs(shap_values), axis=0) > 1]
print(feat)
X = X[feat]
# %%
test_size = 0.33
cv = 3
Xt, Xv, yt, yv = \
sklearn.model_selection.train_test_split(
X, y, test_size=test_size, stratify=y, random_state=seed)
viz = RFECV(XGBClassifier(
max_depth=7,
n_estimators=1000,
scale_pos_weight=50000,
subsample=0.5,
colsample_bylevel=0.5,
colsample_bytree=0.5,
seed=seed,
),
cv=cv)
viz.fit(Xv, yv)
viz.poof()
# %%
feat = feat[viz.ranking_ < 5]
print(feat)
X = X[feat]
# %%
test_size = 0.33
cv = 3