def predict(self, x_test):
if self.verbose > 0:
print("[{}] Predicting data of {}".format(self.name, x_test.shape))
start_time = time.time()
model_path = self.output_directory + 'best_model.hdf5'
model = keras.models.load_model(model_path)
y_pred = model.predict(x_test, batch_size=self.batch_size)
# if return_df_metrics:
# y_pred = np.argmax(y_pred, axis=1)
# df_metrics = calculate_metrics(y_true, y_pred, 0.0)
# return df_metrics
# else:
# test_duration = time.time() - start_time
# save_test_duration(self.output_directory + 'test_duration.csv', test_duration)
# return y_pred
test_duration = time.time() - start_time
save_test_duration(self.output_directory + 'test_duration.csv',
test_duration)
if self.verbose > 0:
print("[{}] Predicting completed, took {}s".format(
self.name, test_duration))
return y_pred, test_duration
def predict(self, x: np.array):
"""
Do prediction using the regression model on x
Inputs:
x: data for prediction (num_examples, num_timestep, num_channels) or (num_examples, num_features)
"""
print("[{}] Predicting data of {}".format(self.name, x.shape))
start_time = time.perf_counter()
if len(x.shape) == 3:
x = x.reshape(x.shape[0], x.shape[1] * x.shape[2])
y_pred = self.model.predict(x)
test_duration = time.perf_counter() - start_time
save_test_duration(self.output_directory + 'test_duration.csv',
test_duration)
print("[{}] Predicting completed, took {}s".format(
self.name, test_duration))
return y_pred
def predict(self, x):
print('[{}] Predicting'.format(self.name))
start_time = time.perf_counter()
print('[{}] Applying kernels'.format(self.name))
x_test_transform = apply_kernels(x, self.kernels)
y_pred = self.regressor.predict(x_test_transform)
test_duration = time.perf_counter() - start_time
save_test_duration(self.output_directory + 'test_duration.csv',
test_duration)
print("[{}] Predicting completed, took {}s".format(
self.name, test_duration))
return y_pred
def predict(self, x):
print('[{}] Predicting'.format(self.name))
start_time = time.perf_counter()
model = tf.keras.models.load_model(self.output_directory +
self.best_model_file)
yhat = model.predict(x)
tf.keras.backend.clear_session()
test_duration = time.perf_counter() - start_time
save_test_duration(self.output_directory + 'test_duration.csv',
test_duration)
print('[{}] Prediction done!'.format(self.name))
return yhat
def predict(self, x_test):
print("[{}] Predicting data of {}".format(self.name, x_test.shape))
start_time = time.perf_counter()
if len(x_test.shape) == 3:
x_test = x_test.reshape(x_test.shape[0], x_test.shape[1] * x_test.shape[2])
y_pred = self.model.predict(x_test)
test_duration = time.perf_counter() - start_time
save_test_duration(self.output_directory + 'test_duration.csv', test_duration)
print("[{}] Predicting completed, took {}s".format(self.name, test_duration))
return y_pred
def predict(self, x_test):
if self.verbose > 0:
print("[{}] Predicting data of {}".format(self.name, x_test.shape))
start_time = time.time()
model_path = self.output_directory + 'best_model.hdf5'
model = keras.models.load_model(model_path)
y_pred = model.predict(x_test)
test_duration = time.time() - start_time
self.test_duration = test_duration
save_test_duration(self.output_directory + 'test_duration.csv', test_duration)
if self.verbose > 0:
print("[{}] Predicting completed, took {}s".format(self.name, test_duration))
return y_pred, test_duration
def predict(self, x_test):
if self.verbose > 0:
print("[{}] Predicting data of {}".format(self.name, x_test.shape))
start_time = time.time()
y_pred = self.model.predict(x_test)
test_duration = time.time() - start_time
save_test_duration(self.output_directory + 'test_duration.csv',
test_duration)
self.test_duration = test_duration
if self.verbose > 0:
print("[{}] Predicting completed, took {}s".format(
self.name, test_duration))
return y_pred, test_duration
def predict(self, x):
"""
Do prediction with DL models
Inputs:
x: data for prediction (num_examples, num_timestep, num_channels)
Outputs:
y_pred: prediction
"""
print('[{}] Predicting'.format(self.name))
start_time = time.perf_counter()
model = tf.keras.models.load_model(self.output_directory + self.best_model_file)
yhat = model.predict(x)
tf.keras.backend.clear_session()
test_duration = time.perf_counter() - start_time
save_test_duration(self.output_directory + 'test_duration.csv', test_duration)
print('[{}] Prediction done!'.format(self.name))
return yhat