def test_f_oneway_ints():
# Smoke test f_oneway on integers: that it does raise casting errors
# with recent numpys
X = np.random.randint(10, size=(10, 10))
y = np.arange(10)
fint, pint = f_oneway(X, y)
# test that is gives the same result as with float
f, p = f_oneway(X.astype(np.float), y)
assert_array_almost_equal(f, fint, decimal=5)
assert_array_almost_equal(p, pint, decimal=5)
def test_f_oneway_ints():
# Smoke test f_oneway on integers: that it does raise casting errors
# with recent numpys
X = np.random.randint(10, size=(10, 10))
y = np.arange(10)
fint, pint = f_oneway(X, y)
# test that is gives the same result as with float
f, p = f_oneway(X.astype(np.float), y)
assert_array_equal(f, fint)
assert_array_equal(p, pint)
def test_f_oneway_vs_scipy_stats():
"""Test that our f_oneway gives the same result as scipy.stats"""
X1 = np.random.randn(10, 3)
X2 = 1 + np.random.randn(10, 3)
f, pv = stats.f_oneway(X1, X2)
f2, pv2 = f_oneway(X1, X2)
assert_true(np.allclose(f, f2))
assert_true(np.allclose(pv, pv2))
def test_f_oneway_vs_scipy_stats():
# Test that our f_oneway gives the same result as scipy.stats
rng = np.random.RandomState(0)
X1 = rng.randn(10, 3)
X2 = 1 + rng.randn(10, 3)
f, pv = stats.f_oneway(X1, X2)
f2, pv2 = f_oneway(X1, X2)
assert_true(np.allclose(f, f2))
assert_true(np.allclose(pv, pv2))
def test_f_oneway_vs_scipy_stats():
# Test that our f_oneway gives the same result as scipy.stats
rng = np.random.RandomState(0)
X1 = rng.randn(10, 3)
X2 = 1 + rng.randn(10, 3)
f, pv = stats.f_oneway(X1, X2)
f2, pv2 = f_oneway(X1, X2)
assert_true(np.allclose(f, f2))
assert_true(np.allclose(pv, pv2))
def test_f_oneway_ints():
# Smoke test f_oneway on integers: that it does raise casting errors
# with recent numpys
f_oneway(np.random.randint(10, size=(10, 10)), np.arange(10))
def test_f_oneway_ints():
# Smoke test f_oneway on integers: that it does raise casting errors
# with recent numpys
f_oneway(np.random.randint(10, size=(10, 10)), np.arange(10))
import metric as mm
LabelEncoder=preprocessing.LabelEncoder().fit_transform
#将变量分为因子变量和数值变量分开讨论
label_numeric=np.array([np.issubdtype(X.dtypes[i],np.number) for i in range(X.shape[1])])
fscore_nonumeric=X.loc[:,~label_numeric].describe().T.assign(missing_pct=lambda x:(N-x['count'])/N)
fscore_numeric=X.loc[:,label_numeric].describe().T.assign(missing_pct=lambda x:(N-x['count'])/N)
for cc in fscore_nonumeric.index:
fscore_nonumeric.loc[cc,'iv']=mm.info_value(X[cc],Y)
f,p=fs.f_oneway(X.loc[:,label_numeric],Y)
fscore_numeric['f_oneway']=f
fscore_numeric['f_oneway_p']=p
# 测试数据
training=np.random.choice([True,False],p=[0.8,0.2],size=N)
# 建模
#X.loc[:,~label_numeric]=X.loc[:,~label_numeric].apply(LabelEncoder)
X1=pd.get_dummies(X.loc[:,~label_numeric],drop_first=True)
X2=pd.concat([X1,X.loc[:,label_numeric]],axis=1)
