Y_DIM = 5
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('-X', help = 'X.h5 file (with path)')
parser.add_argument('--Y', help = '(optional) Y.h5 file (with path)')
parser.add_argument('--file_out', help = '(optional) .h5 file output (with path)')
args = parser.parse_args()
# reading X data
filenameX = args.X
r = reader.H5Reader({'X':(None,X_DIM), 'Y':(None,Y_DIM)})
dataX = r.read(filenameX)
X = dataX['X']
# load models
filename = os.path.join( PATH_MDL, 'neural_network.sav')
model = joblib.load( filename )
filename = os.path.join( PATH_MDL, 'scalerX.sav')
scalerX = joblib.load( filename )
filename = os.path.join( PATH_MDL, 'scalerY.sav')
scalerY = joblib.load( filename )
filename = os.path.join( PATH_MDL, 'linreg.sav')
benchmark = joblib.load( filename )
# neural_network predict
import reader
import os
import tools
import liquid_algos as algo
import joblib
path_data = 'data'
path_mdl = 'model'
if __name__ == '__main__':
### read data
filenameX = os.path.join(path_data, 'X.h5')
filenameY = os.path.join(path_data, 'Y.h5')
r = reader.H5Reader(data_shape={'Y': (None, 5)})
dataX = r.read(filenameX)
dataY = r.read(filenameY)
X = dataX['X']
Y = dataY['Y']
# train NN
scalerX, scalerY = algo.fit_scaler(X), algo.fit_scaler(Y)
Xscale, Yscale = scalerX.transform(X), scalerY.transform(Y)
model = algo.sklearn_nn(hidden_layer_sizes=(256, 128),
activation='relu',
max_iter=100)
model.fit(Xscale, Yscale)