def test():
lvl = 1
wavelet = 'db4' # Haar'
ts_file_name = 'ford_ts.csv'
last_days = 1200
time_frame = 60
time_bias = 1
data_loader = DataLoader(ts_file_name, last_days, debug=True)
raw_data = data_loader.as_matrix()
ts_data = denoise(raw_data, lvl, wavelet)
# plt.plot(raw_data[3])
# plt.show()
# plt.plot(ts_data[3])
# plt.show()
daily_features, _ = np.shape(ts_data)
dataset = data_loader.prepare_dataset_sae(ts_data, time_frame, time_bias)
runner = Runner(daily_features,
lstm_layers=1,
gamma=0.005,
delay=4,
sae_lr=0.01,
beta=0,
hidden_nodes_activation_rate=0.9,
hidden_layers_sizes=[8],
debug=True)
cross_validator = CrossValidator()
pred_target = cross_validator.run_validation(runner,
dataset,
sae_epoch=1,
lstm_epoch=1)
pred_target_dollars = [(data_loader.to_dolar(x), data_loader.to_dolar(y))
for x, y in pred_target]
dollars_loss = sum([abs(x - y) for x, y in pred_target_dollars])
print("[RUNNER] Dollars lost={}".format(dollars_loss))
def test_final(with_beta=False):
lvl = 1
wavelet = 'db4' # Haar'
ts_file_name = 'ford_final.csv'
training_days = 1200
days_test_frame = 90
tests_number = 6
time_frame = 60
time_bias = 1
sae_epoch = 200
lstm_epoch = 50
if with_beta:
beta = 1
print("[TEST_FINAL] EVALUATION OF MODEL WITH BETA=1")
else:
beta = 0
print("[TEST_FINAL] EVALUATION OF MODEL WITH BETA=0")
all_days = training_days + days_test_frame * tests_number + 30 + time_frame
data_loader = DataLoader(ts_file_name, all_days, debug=True)
raw_data = data_loader.as_matrix()
ts_data = denoise(raw_data, lvl, wavelet)
daily_features, _ = np.shape(ts_data)
dataset = data_loader.prepare_dataset_sae(ts_data, time_frame, time_bias)
runner = Runner(daily_features,
lstm_layers=1,
gamma=0.005,
delay=4,
sae_lr=0.01,
beta=beta,
hidden_nodes_activation_rate=0.9,
hidden_layers_sizes=[8],
debug=True)
validator = Validator()
training_dataset = dataset
test_datasets = []
for _ in range(tests_number):
ntest = training_dataset[-days_test_frame:]
training_dataset = training_dataset[:-days_test_frame] #TODO CHECK!
test_datasets.append(ntest)
test_datasets.reverse()
pred_target = []
for test_dataset in test_datasets:
pred_target += validator.train_and_eval_hard(runner, training_dataset,
test_dataset, sae_epoch,
lstm_epoch)
training_dataset += test_dataset
pred_target_dollars = [(data_loader.to_dolar(x), data_loader.to_dolar(y))
for x, y in pred_target]
dollars_loss = sum([abs(x - y) for x, y in pred_target_dollars])
print("[RUNNER] Total dollars lost={}".format(dollars_loss))
print("==============================================================")
