parser.add_argument("savefile", help="path to saved .mat file", type=str)
args = parser.parse_args()
############################################################################
# Read config file
############################################################################
config = json.load(open(args.esnconfig + '.json', 'r'))
############################################################################
# Load data
############################################################################
# If the data is stored in a directory, load the data from there. Otherwise,
# load from the single file and split it.
if os.path.isdir(args.data):
Xtr, Ytr, _, _, Xte, Yte = esnet.load_from_dir(args.data)
else:
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, _, _, Xte, Yte = esnet.generate_datasets(X, Y)
def main():
# Run in parallel and store result in a numpy array
Yhat,error,train_states,train_embedding,test_states,test_embedding = esnet.run_from_config_return_states(Xtr, Ytr, Xte, Yte, config)
savemat(args.savefile, {'train_states':train_states, 'train_embedding':train_embedding, 'test_states':test_states, 'test_embedding':test_embedding})
if __name__ == "__main__":
main()
elif dataType=='SantaFe' or dataType=='Sunspots' or dataType=='Hongik' \
or dataType=='GEFC' or dataType=='Mackey' or dataType=='SP500' \
or dataType=='Rainfall' or dataType=='Temperature' \
or dataType == 'MinTempMel' or dataType == 'SunSpotsZu'\
or dataType == 'TempAlbuquerque' or dataType == 'TempDenver' or dataType == 'TempLasVegas' \
or dataType == 'TempLosAngeles' or dataType == 'TempPhoenix' or dataType == 'TempPortland' \
or dataType == 'TempSanDiego' or dataType == 'TempSanFrancisco' or dataType == 'TempSeattle' \
or dataType == 'TempVancouver' \
or dataType == 'eleGB2015_7_12' or dataType == 'eleDE2015_7_12' or dataType == 'eleFR2015_7_12'\
or dataType == 'Electric':
#Xtr, Ytr, _, _, Xte, Yte, Yscaler = esnet.generate_datasets_santafe(args.data)
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, _, _, Xte, Yte, Yscaler = esnet.generate_datasets(X, Y)
Xtr, Xte = esnet.reconstruct_input_1d([Xtr, Xte], reconstructconfig)
Ytr, Yte = esnet.reconstruct_output_1d([Ytr, Yte], reconstructconfig)
elif dataType == 'GEFC_temp' or dataType == 'HenonMap':
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, _, _, Xte, Yte, Yscaler = esnet.generate_datasets(X, Y)
# Reconstruct
Xtr, Xte = esnet.reconstruct_input_2d([Xtr, Xte], reconstructconfig)
Ytr, Yte = esnet.reconstruct_output_2d([Ytr, Yte], reconstructconfig)
else:
optconfig = paramhelper._optimization
############################################################################
# Load data
############################################################################
print("Loading data (%s)" % args.data)
# If the data is stored in a directory, load the data from there. Otherwise,
# load from the single file and split it.
if os.path.isdir(args.data):
Xtr, Ytr, Xval, Yval, _, _ = esnet.load_from_dir(args.data)
else:
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, Xval, Yval, _, _ = esnet.generate_datasets(X, Y)
############################################################################
# Initialization of the genetic algorithm
############################################################################
# Fitness and individual. Different formats, depending on dimensionality reduction.
if paramhelper._fixed_values['embedding'] == 'identity':
creator.create("FitnessMin", base.Fitness,
weights=(-1.0, )) # -1.0 => minimize function
else:
creator.create("FitnessMin", base.Fitness,
weights=(-1.0, -0.1)) # -1.0 => minimize function
# The individuals are dicts of numbers (parameters)
# The length and the type of number varies across ESN configurations.
creator.create("Individual", dict, fitness=creator.FitnessMin)
############################################################################
# Load data
############################################################################
logger.info("Loading data (%s)" % args.data)
# If the data is stored in a directory, load the data from there. Otherwise,
# load from the single file and split it.
if os.path.isdir(args.data):
Xtr, Ytr, Xval, Yval, _, _, Yscaler = esnet.load_from_dir(args.data)
else:
count = args.count
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, Xval, Yval, _, _, Yscaler = esnet.generate_datasets(
X[:count], Y[:count])
############################################################################
# Initialization of the genetic algorithm
############################################################################
# Fitness and individual. Different formats, depending on dimensionality reduction.
if paramhelper._fixed_values['embedding'] == 'identity':
creator.create("FitnessMin", base.Fitness,
weights=(-1.0, )) # -1.0 => minimize function
else:
creator.create("FitnessMin", base.Fitness,
weights=(-1.0, -0.1)) # -1.0 => minimize function
# The individuals are dicts of numbers (parameters)
# The length and the type of number varies across ESN configurations.
creator.create("Individual", dict, fitness=creator.FitnessMin)
elif dataType=='SantaFe' or dataType=='Sunspots' or dataType=='Hongik' \
or dataType=='GEFC' or dataType=='Mackey' or dataType=='SP500' \
or dataType == 'Rainfall' or dataType=='Temperature'\
or dataType=='MinTempMel' or dataType=='SunSpotsZu' \
or dataType == 'TempAlbuquerque' or dataType == 'TempDenver' or dataType=='TempLasVegas' \
or dataType == 'TempLosAngeles' or dataType == 'TempPhoenix' or dataType == 'TempPortland' \
or dataType == 'TempSanDiego' or dataType == 'TempSanFrancisco' or dataType == 'TempSeattle'\
or dataType == 'TempVancouver' \
or dataType == 'eleGB2015_7_12' or dataType == 'eleDE2015_7_12' or dataType == 'eleFR2015_7_12' \
or dataType == 'Electric':
#Xtr, Ytr, Xval, Yval, _, _, Yscaler = esnet.generate_datasets_santafe(args.data)
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, Xval, Yval, _, _, Yscaler = esnet.generate_datasets(X, Y)
Xtr, Xval = esnet.reconstruct_input_1d([Xtr, Xval], reconstructconfig)
Ytr, Yval = esnet.reconstruct_output_1d([Ytr, Yval], reconstructconfig)
elif dataType=='GEFC_temp' or dataType=='HenonMap':
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, Xval, Yval, _, _, Yscaler = esnet.generate_datasets(X, Y)
# Reconstruct
Xtr, Xval = esnet.reconstruct_input_2d([Xtr, Xval], reconstructconfig)
Ytr, Yval = esnet.reconstruct_output_2d([Ytr, Yval], reconstructconfig)
else:
############################################################################
# Load data
############################################################################
logger.info("Loading data (%s)" % args.data)
# If the data is stored in a directory, load the data from there. Otherwise,
# load from the single file and split it.
if os.path.isdir(args.data):
Xtr, Ytr, Xval, Yval, _, _ = esnet.load_from_dir(args.data)
else:
X, Y = esnet.load_from_text(args.data)
# Construct training/test sets
Xtr, Ytr, Xval, Yval, _, _ = esnet.generate_datasets(X,
Y,
scaler=MinMaxScaler)
############################################################################
# Initialization of the genetic algorithm
############################################################################
# Fitness and individual. Different formats, depending on dimensionality reduction.
if paramhelper._fixed_values['embedding'] == 'identity':
creator.create("FitnessMin", base.Fitness,
weights=(-1.0, )) # -1.0 => minimize function
else:
creator.create("FitnessMin", base.Fitness,
weights=(-1.0, -0.1)) # -1.0 => minimize function
# The individuals are dicts of numbers (parameters)
# The length and the type of number varies across ESN configurations.
def main():
averages = []
predictions_error = []
predictions = []
reals = []
startPoint = args.count
#For the first few predictions, use the last error as prediction
dataPath = '/home/minh/PycharmProjects/Ensemble/PythonESN/data/edgar_historical'
count = 0
"""with open(dataPath,'r') as f:
for line in f:
if count>0:
data = line.split(',')
total = 0
for i in range(5):
total += float(data[i])
averages.append(total/5)
reals.append(float(data[6]))
count+=1"""
"""count = 0
with open(dataPath,'r') as f:
for line in f:
if count>0:
if count>(startPoint+2):
break
else:
data = line.split(',')
total = 0
for i in range(5):
total += float(data[i])
predictions.append(float(data[6])-total/5)
count+=1
print('predictions:',predictions)"""
# Run in parallel and store result in a numpy array
X, Y = esnet.load_from_text(args.data)
Xtr, Ytr, _, _, Xte, Yte, Yscaler = esnet.generate_datasets(X[:startPoint], Y[:startPoint])
Yhat, error = esnet.run_from_config(Xtr, Ytr, Xte, Yte, config, Yscaler)
Yhat = np.ceil(Yscaler.inverse_transform(Yhat))
predictions_error.append(Yhat[len(Yhat) - 1][0])
"""for i in range(startPoint,len(X)):
Xtr, Ytr, _, _, Xte, Yte, Yscaler = esnet.generate_datasets(X[:i], Y[:i])
if i<800:
config['n_drop'] = int(i/8)
else:
config['n_drop'] = 100
Yhat, error = esnet.run_from_config(Xtr, Ytr, Xte, Yte, config)
Yhat = np.ceil(Yscaler.inverse_transform(Yhat))
#print('predictions:',Yhat)
#print('error:',error)
predictions_error.append(Yhat[len(Yhat)-1][0])
count+=1
if count%100==0:
print('predictions made:',count)"""
curPath = os.getcwd().split('/')
writePath = ''
for i in range(len(curPath)):
writePath += curPath[i] + '/'
configs = args.esnconfig.split('/')
writePath += 'predictions/predictions_' + configs[-1] + '_' + str(args.times)
"""with open(writePath, 'a') as f:
f.write(str(np.ceil(Yhat[len(Yhat) - 1][0]))+'\n')"""
with open (writePath, 'w') as f:
for i in range(len(Yhat)):
f.writelines(str(np.ceil(Yhat[i][0])) + '\n')