class Manager:
provider = Selector()
def nn(self, test_no):
nn = Nn(self.provider)
return nn.test(test_no)
def linear(self, test_no):
linear = Linear(self.provider)
return linear.test(test_no)
def tree(self, test_no):
tree = Rt(self.provider)
return tree.test(test_no)
def svr(self, test_no):
svr = SvrN(self.provider)
return svr.test(test_no)
from provider import Provider
from feature_selection import Selector
from dimension_reduction import Reducer
from sklearn import tree
from sklearn.metrics import mean_squared_error
from math import sqrt
import matplotlib.pyplot as plt
if __name__ == '__main__':
provider = Selector()
input = []
target = []
for d in provider.getLearnData():
input.append(d[0])
target.append(d[1][0])
regressor = tree.DecisionTreeRegressor()
regressor.fit(input, target)
mult = 1 #provider.multiplier
results = regressor.predict(input)
print(results)
w = open('output4_1_sr.txt', 'w')
for i, line in enumerate(results):
w.write('%f;%f\n' % (target[i] * mult, line * mult))
w.close()
input = []
target = []
etrue = []
epredict = []
from provider import Provider
from feature_selection import Selector
import numpy as np
import neurolab as nl
from sklearn.metrics import mean_squared_error
from math import sqrt
if __name__ == '__main__':
provider = Selector()
net = nl.net.newff(provider.getDataRanges(), [provider.getInputCount()*2, provider.getInputCount()*2, 1])
# net.trainf = nl.train.train_gd
input = []
target = []
mult = provider.provider.multiplier
for d in provider.getLearnData():
input.append(d[0])
target.append([d[1][0]/mult])
err = net.train(input, target, epochs=1000, show=10, goal=0.001)#, lr=0.000001)
results = net.sim(input)
w = open('output1.txt', 'w')
for i, line in enumerate(results):
w.write('%f;%f\n' % (target[i][0]*mult, line[0]*mult))
w.close()
input = []
target = []
for d in provider.getTestData():
input.append(d[0])
target.append([d[1][0]/mult])
results = net.sim(input)
from provider import Provider
from feature_selection import Selector
import numpy as np
import neurolab as nl
from sklearn.metrics import mean_squared_error
from math import sqrt
import matplotlib.pyplot as plt
if __name__ == '__main__':
provider = Selector()
net = nl.net.newff(
provider.getDataRanges(),
[provider.getInputCount() * 2,
provider.getInputCount() * 2, 1])
# net.trainf = nl.train.train_gd
input = []
target = []
mult = provider.provider.multiplier
for d in provider.getLearnData():
input.append(d[0])
target.append([d[1][0] / mult])
err = net.train(input, target, epochs=1000, show=10,
goal=0.001) #, lr=0.000001)
results = net.sim(input)
w = open('output1.txt', 'w')
for i, line in enumerate(results):
w.write('%f;%f\n' % (target[i][0] * mult, line[0] * mult))
w.close()
input = []
from feature_selection import Selector
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
from math import sqrt
import matplotlib.pyplot as plt
if __name__ == '__main__':
# reducer = Reducer()
reducer = Selector()
input = []
target = []
for d in reducer.getLearnData():
input.append(d[0])
target.append(d[1][0])
regressor = LinearRegression()
regressor.fit(input, target)
print(regressor.coef_)
mult = 1 #reducer.provider.multiplier
results = regressor.predict(input)
print(results)
w = open('output2_1_s.txt', 'w')
for i, line in enumerate(results):
w.write('%f;%f\n' % (target[i] * mult, line * mult))
w.close()
input = []
target = []
for d in reducer.getTestData():