def test_grafic_sensibility(self):
INSTANCE: int = 74
x, y = make_moons(n_samples=1500, noise=.4, random_state=17)
clf = MLPClassifier()
x_train, x_test, y_train, y_test = train_test_split(x,
y,
train_size=.8,
test_size=.2,
random_state=17)
clf.fit(x_train, y_train)
gpx = Gpx(clf.predict_proba,
x_train=x,
y_train=y,
random_state=42,
feature_names=['x', 'y'])
gpx.explaining(x_test[INSTANCE, :])
x, y = gpx.x_around[:, 0], gpx.x_around[:, 1]
y_proba = gpx.proba_transform(gpx.y_around)
resolution = 0.02
x1_min, x1_max = x.min() - 1, x.max() + 1
x2_min, x2_max = y.min() - 1, y.max() + 1
xm1, xm2 = np.meshgrid(np.arange(x1_min, x1_max, resolution),
np.arange(x2_min, x2_max, resolution))
Z_bb = gpx.gp_prediction(np.array([xm1.ravel(), xm2.ravel()]).T)
fig, ax = plt.subplots()
ax.set_xlim(x1_min, x1_max)
ax.set_xlim(x2_min, x2_max)
scat = plt.scatter(x, y, y_proba)
def func(data):
k, j = data
scat.set_offsets(k)
scat.set_array(j)
mmm = gpx.max_min_matrix(noise_range=10)
gen = []
for n in mmm[:, 0]:
aux = gpx.x_around.copy()
aux[:, 0] = n
gen.append((aux.copy(), gpx.gp_prediction(aux.copy())))
animation = ani.FuncAnimation(fig,
func,
gen,
interval=200,
save_count=200)
plt.contourf(xm1, xm2, Z_bb.reshape(xm1.shape), alpha=0.4)
plt.scatter(x, y, c=y_proba)
plt.show()
writergif = ani.PillowWriter(fps=5)
animation.save('sens_x_2.gif', writer=writergif)
sens_gpx = gpx.feature_sensitivity()
print(sens_gpx)
def test_multi_class(self):
INSTANCE: int = 20
iris = load_iris()
X, y = load_iris(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=.3,
random_state=42)
scaler = Normalizer()
scaler.fit(X_train)
X_train = scaler.transform(X_train)
clf = RandomForestClassifier(random_state=42)
clf.fit(X_train, y_train)
gp_hyper_parameters = {
'population_size':
100,
'generations':
100,
'stopping_criteria':
0.0001,
'p_crossover':
0.7,
'p_subtree_mutation':
0.1,
'p_hoist_mutation':
0.05,
'p_point_mutation':
0.1,
'const_range': (-1, 1),
'parsimony_coefficient':
0.0005,
'init_depth': (3, 6),
'n_jobs':
-1,
'random_state':
42,
'function_set': ('add', 'sub', 'mul', 'div', 'sqrt', 'log', 'abs',
'neg', 'inv', 'max', 'min', 'sin', 'cos', 'tan')
}
gpx = Gpx(clf.predict_proba,
gp_hyper_parameters=gp_hyper_parameters,
x_train=X_train,
y_train=y_train,
feature_names=iris.feature_names,
num_samples=500,
k_neighbor=50,
random_state=42)
y_hat = gpx.explaining(
scaler.transform(X_test[INSTANCE, :].reshape(-1, 1)))
x_around = gpx.x_around
gpx_y = gpx.gp_prediction(x_around)
bb_y = clf.predict(x_around)
gpx.logger.info('Multiclass gpx_y:{} / bb_y {}'.format(gpx_y, bb_y))
gpx.logger.info('test_understand mult-class accuracy {}'.format(
accuracy_score(gpx_y, bb_y)))