def train(self, k, train_set, valid_set):
train_wavs, train_folds, train_labels = zip(*list(chain(*train_set)))
train_wavs, train_folds, train_labels = np.array(train_wavs), np.array(
train_folds), np.array(train_labels)
train_sample = len(train_wavs)
train_x, _ = self.fix_frame(train_sample, train_wavs, train_folds,
train_labels)
# Test Model
valid_wavs, valid_folds, valid_labels = zip(*valid_set)
valid_wavs, valid_folds, valid_labels = np.array(valid_wavs), np.array(
valid_folds), np.array(valid_labels)
valid_sample = len(valid_wavs)
valid_x, valid_y = self.fix_frame(valid_sample, valid_wavs,
valid_folds, valid_labels)
if config.isPCA:
pca = PCA(n_components=config.n_pca)
pca.fit(train_x)
train_x = pca.transform(train_x)
valid_x = pca.transform(valid_x)
kmeans = KMeans(n_clusters=self.n_class, random_state=0).fit(train_x)
joblib.dump(kmeans, f"{self.model_path}/kmeans-{k}.pkl")
score = purity_score(np.argmax(valid_y, axis=1),
kmeans.predict(valid_x))
print('Accuracy:{0:.3f}'.format(score * 100))
def train(self, k, train_set, valid_set):
train_wavs, train_folds, train_labels = zip(*list(chain(*train_set)))
train_wavs, train_folds, train_labels = np.array(train_wavs), np.array(train_folds), np.array(train_labels)
train_sample = len(train_wavs)
train_x, _ = self.fix_frame(train_sample, train_wavs, train_folds, train_labels)
# Test Model
valid_wavs, valid_folds, valid_labels = zip(*valid_set)
valid_wavs, valid_folds, valid_labels = np.array(valid_wavs), np.array(valid_folds), np.array(valid_labels)
valid_sample = len(valid_wavs)
valid_x, valid_y = self.fix_frame(valid_sample, valid_wavs, valid_folds, valid_labels)
if config.isPCA:
pca = PCA(n_components=config.n_pca)
pca.fit(train_x)
train_x = pca.transform(train_x)
valid_x = pca.transform(valid_x)
gmm = GaussianMixture(n_components=self.component).fit(train_x)
joblib.dump(gmm, f"{self.model_path}/gmm-{k}.pkl")
score = purity_score(np.argmax(valid_y, axis=1), gmm.predict(valid_x))
print('Accuracy:{0:.3f}'.format(score))
def test(self, k, test_set):
# Test Model
load_model = joblib.load(f"{self.model_path}/gmm-{k}.pkl")
test_wavs, test_folds, test_labels = zip(*test_set)
test_wavs, test_folds, test_labels = np.array(test_wavs), np.array(test_folds), np.array(test_labels)
test_samples = len(test_wavs)
test_x, test_y = self.fix_frame(test_samples, test_wavs, test_folds, test_labels)
score = purity_score(np.argmax(test_y, axis=1), load_model.predict(test_x))
return score