def setUp(self):
np.random.seed(31337)
self.project_path = get_project_path()
self.data_dir = join(self.project_path, 'crypr', 'tests', 'data')
self.SYM = 'ETH'
self.LAST_N_HOURS = 14000
self.FEATURE_WINDOW = 72
self.MOVING_AVERAGE_LAGS = [6, 12, 24, 48, 72]
self.TARGET = 'close'
self.Tx = 72
self.Ty = 1
self.TEST_SIZE = 0.05
self.end_time = utc_timestamp_ymd(2018, 6, 27)
self.data = retrieve_all_data(coin=self.SYM,
num_hours=self.LAST_N_HOURS,
comparison_symbol='USD',
end_time=self.end_time)
self.predict_data = retrieve_all_data(coin=self.SYM,
num_hours=self.Tx +
self.FEATURE_WINDOW - 1,
comparison_symbol='USD',
end_time=self.end_time)
self.X_shape = (13852, 1224)
self.y_shape = (13852, 1)
self.X_sample = 709.48
self.y_sample = -1.498064809896027
self.X_train_shape = (13159, 1224)
self.X_test_shape = (693, 1224)
self.y_train_shape = (13159, 1)
self.y_test_shape = (693, 1)
self.X_train_sample = 11.41
self.y_train_sample = 0.0
self.X_test_sample = 487.58
self.y_test_sample = 0.9448599618077758
self.parameters = {
'objective': 'reg:linear',
'learning_rate': .07,
'max_depth': 10,
'min_child_weight': 4,
'silent': 1,
'subsample': 0.7,
'colsample_bytree': 0.7,
'n_estimators': 20,
}
self.train_mae = 0.8953377462440475
self.train_rmse = 1.4144230033451395
self.prediction = 1.2296733856201172
def runTest(self):
for hours in self.num_hours:
data = retrieve_all_data(coin=self.coin,
num_hours=hours,
comparison_symbol=self.comparison_sym,
exchange=self.exchange,
end_time=self.end_to_time)
self.shapeCheck(data=data, num_hours=hours)
self.columnsCheck(data=data)
self.ascendingUniqueCheck(data=data)
self.equalSpacingCheck(data=data)
def predict(coin=None):
coin = coin or 'BTC'
wavelet = 'haar'
Tx = 72
Ty = 1
target = 'close'
cryptocompare_data = retrieve_all_data(coin, Tx + 1)
preprocessor = DWTSmoothPreprocessor(
production=True,
target_col=target,
Tx=Tx,
Ty=Ty,
wavelet=wavelet,
name='CryptoPredict_DWTSmoothPreprocessor_{}'.format(coin))
preprocessed_data = preprocessor.fit(cryptocompare_data).transform(
cryptocompare_data)
if coin == 'ETH':
model = eth_model
elif coin == 'BTC':
model = btc_model
else:
#FIXME: More descriptive error
abort(404)
with model.graph.as_default():
prediction = model.estimator.predict(preprocessed_data)
def parse_keras_prediction(keras_prediction):
"""Handles multi-output Keras models"""
if len(keras_prediction) == 2:
return keras_prediction[1][0]
else:
return keras_prediction[0]
parsed_prediction = parse_keras_prediction(prediction)
last_value = cryptocompare_data[target].iloc[-1]
last_time = cryptocompare_data['timestamp'].iloc[-1]
prediction_val = [
last_value + pred / 100 * last_value for pred in parsed_prediction
]
time_val = [
last_time + pd.Timedelta(hours=1 * (int(ix) + 1))
for ix in range(len(parsed_prediction))
]
return dict(prediction=prediction_val, time=time_val)
def main(hours):
print('Downloading data from Cryptocompare ...')
output_dir = join(get_project_path(), 'data', 'raw')
makedirs(output_dir, exist_ok=True)
coins = ['BTC', 'ETH']
for coin in coins:
print('Retrieving {} coin data from API...'.format(coin))
raw_df = retrieve_all_data(coin=coin,
num_hours=hours,
comparison_symbol='USD')
output_path = join(output_dir, coin + '.csv')
raw_df.to_csv(output_path)
def main(hours):
log_fmt = '%(asctime)s - %(name)s - %(levelname)s - %(message)s'
logging.basicConfig(level=logging.INFO, format=log_fmt)
load_dotenv(find_dotenv())
project_path = os.path.dirname(find_dotenv())
logger = logging.getLogger(__name__)
logger.info('Downloading data from Cryptocompare ...')
output_path = '{}/data/raw'.format(project_path)
if not os.path.exists(output_path):
print('Making output directory...')
os.makedirs(output_path)
coins = ['BTC', 'ETH']
for coin in coins:
coin_data = cryptocompare.retrieve_all_data(coin=coin,
num_hours=hours,
comparison_symbol='USD')
coin_output_path = '{}/{}.csv'.format(output_path, coin)
coin_data.to_csv(coin_output_path)
def predict(coin=None):
coin = coin or 'BTC'
Tx = 72
target = 'close'
cryptocompare_data = retrieve_all_data(coin, Tx + 1 + 48)
preprocessed_data = make_features(cryptocompare_data, target_col=target,
keep_cols=['close', 'high', 'low', 'volumeto', 'volumefrom'],
ma_lags=[6, 12, 24, 48], ma_cols=['close', 'volumefrom', 'volumeto'])
time_series_data = series_to_predict_matrix(preprocessed_data, Tx)
n_features = int(time_series_data.shape[1]/Tx)
model_input_data = make_3d(arr=time_series_data, tx=Tx, num_channels=n_features)
if coin == 'ETH':
model = eth_model
elif coin == 'BTC':
model = btc_model
else:
abort(404) # FIXME: More descriptive error
with model.graph.as_default():
prediction = model.estimator.predict(model_input_data)
def parse_keras_prediction(keras_prediction):
"""Handles multi-output Keras models"""
if len(keras_prediction) == 2:
return keras_prediction[1][0]
else:
return keras_prediction[0]
parsed_prediction = parse_keras_prediction(prediction)
last_value = cryptocompare_data[target].iloc[-1]
last_time = cryptocompare_data['timestamp'].iloc[-1]
prediction_val = [last_value + pred/100*last_value for pred in parsed_prediction]
time_val = [last_time + pd.Timedelta(hours=ix + 1) for ix in range(len(parsed_prediction))]
return dict(prediction=prediction_val, time=time_val)
SYM = 'ETH'
LAST_N_HOURS = 14000
FEATURE_WINDOW = 72
MOVING_AVERAGE_LAGS = [6, 12, 24, 48, 72]
TO_TIME = utc_timestamp_ymd(2018, 6, 27)
TARGET = 'close'
Tx = 72
Ty = 1
TEST_SIZE = 0.05
project_path = get_project_path()
test_data_dir = join(project_path, 'crypr', 'tests', 'data')
data = retrieve_all_data(coin=SYM,
num_hours=LAST_N_HOURS,
comparison_symbol='USD',
end_time=TO_TIME)
preprocessor = SimplePreprocessor(production=False,
target_col=TARGET,
Tx=Tx,
Ty=Ty,
moving_averages=MOVING_AVERAGE_LAGS)
X, y = preprocessor.fit(data).transform(data)
old_shape = X.shape
new_shape = (old_shape[0], old_shape[1] * old_shape[2])
X = pd.DataFrame(np.reshape(a=X, newshape=new_shape),
columns=preprocessor.engineered_columns)
print('X shape: {}'.format(X.shape))
SYM = 'ETH'
LAST_N_HOURS = 14000
FEATURE_WINDOW=72
MOVING_AVERAGE_LAGS = [6, 12, 24, 48, 72]
TARGET = 'close'
Tx = 72
Ty = 1
TEST_SIZE = 0.05
load_dotenv(find_dotenv())
project_path = os.path.dirname(find_dotenv())
data = retrieve_all_data(
coin=SYM,
num_hours=LAST_N_HOURS,
comparison_symbol='USD',
end_time=(np.datetime64(datetime.datetime(2018, 6, 27)).astype('uint64') / 1e6).astype('uint32'))
preprocessor = SimplePreprocessor(
production=False,
target_col=TARGET,
Tx=Tx,
Ty=Ty,
moving_averages=MOVING_AVERAGE_LAGS,
name='unit_test')
X, y = preprocessor.fit(data).transform(data)
old_shape = X.shape
new_shape = (old_shape[0], old_shape[1] * old_shape[2])
X = pd.DataFrame(np.reshape(a=X, newshape=new_shape), columns=preprocessor.engineered_columns)
