def test_Kmeans_init():
"""Test Kmeans"""
covset = generate_cov(20, 3)
labels = np.array([0, 1]).repeat(10)
# init
km = Kmeans(2)
# fit
km.fit(covset)
# fit with init
km = Kmeans(2, init=covset[0:2])
km.fit(covset)
# fit with labels
km.fit(covset, y=labels)
# predict
km.predict(covset)
# transform
km.transform(covset)
# n_jobs
km = Kmeans(2, n_jobs=2)
km.fit(covset)
def test_Kmeans_init():
"""Test Kmeans"""
covset = generate_cov(20, 3)
labels = np.array([0, 1]).repeat(10)
# init
km = Kmeans(2)
# fit
km.fit(covset)
# fit with init
km = Kmeans(2, init=covset[0:2])
km.fit(covset)
# fit with labels
km.fit(covset, y=labels)
# predict
km.predict(covset)
# transform
km.transform(covset)
# n_jobs
km = Kmeans(2, n_jobs=2)
km.fit(covset)
def test_Kmeans_transform():
"""Test transform of Kmeans"""
covset = generate_cov(20, 3)
km = Kmeans(2)
km.fit(covset)
km.transform(covset)
def test_Kmeans_predict():
"""Test prediction of Kmeans"""
covset = generate_cov(20, 3)
km = Kmeans(2)
km.fit(covset)
km.predict(covset)
def test_Kmeans_fit_parallel():
"""Test Fit of Kmeans using paralell"""
covset = generate_cov(20, 3)
km = Kmeans(2, n_jobs=2)
km.fit(covset)
def test_Kmeans_fit_with_y():
"""Test Fit of Kmeans with a given y"""
covset = generate_cov(20, 3)
labels = np.array([0, 1]).repeat(10)
km = Kmeans(2)
km.fit(covset, y=labels)
def test_Kmeans_fit_with_init():
"""Test Fit of Kmeans wit matric initialization"""
covset = generate_cov(20, 3)
km = Kmeans(2, init=covset[0:2])
km.fit(covset)
def test_Kmeans_fit():
"""Test Fit of Kmeans"""
covset = generate_cov(20, 3)
km = Kmeans(2)
km.fit(covset)
filtered_signal = np.expand_dims(filtered_signal, axis=0)
epochs_data = np.concatenate((epochs_data, filtered_signal), axis=0)
## No Filtering
# raw_evoked_signal = evoked.data
# raw_evoked_signal = np.array(raw_evoked_signal)
# raw_evoked_signal = np.expand_dims(raw_evoked_signal, axis=0)
cov_ext_trials = Covariances(estimator='lwf').transform(epochs_data)
### TREINO ONLINE: AQUI NÃO FAZ SENTIDO
#labels = np.append(labels, labels[count])
# Fit a cada 4 janelas de tempo
if (count % 4 == 0 and count != 0 and retrain == True):
kmeans.fit(cov_ext_trials)
print("RETRAINED")
if (count > 5):
prediction_labeled = kmeans.predict(cov_ext_trials)
else:
prediction_labeled = mdm.predict(cov_ext_trials)
# Finish Time Counter
time_2 = time.time()
time_array.append(time_2 - time_1)
count += 1
print("Predictions: ")
print(prediction_labeled[:32])
print(prediction_labeled[32:])
# print ("Label: " + str(labels[i]))
def test_Kmeans_transform():
"""Test transform of Kmeans"""
covset = generate_cov(20,3)
km = Kmeans(2)
km.fit(covset)
km.transform(covset)
def test_Kmeans_predict():
"""Test prediction of Kmeans"""
covset = generate_cov(20,3)
km = Kmeans(2)
km.fit(covset)
km.predict(covset)
def test_Kmeans_fit_parallel():
"""Test Fit of Kmeans using paralell"""
covset = generate_cov(20,3)
km = Kmeans(2,n_jobs=2)
km.fit(covset)
def test_Kmeans_fit_with_y():
"""Test Fit of Kmeans with a given y"""
covset = generate_cov(20,3)
labels = np.array([0,1]).repeat(10)
km = Kmeans(2)
km.fit(covset,y=labels)
def test_Kmeans_fit_with_init():
"""Test Fit of Kmeans wit matric initialization"""
covset = generate_cov(20,3)
km = Kmeans(2,init=covset[0:2])
km.fit(covset)
def test_Kmeans_fit():
"""Test Fit of Kmeans"""
covset = generate_cov(20,3)
km = Kmeans(2)
km.fit(covset)