def run_with_crossvalidation(self, data, iterations=5):
"""
This function estimates the performance of the neural network using crossvalidation using a specified dataset.
Args:
:param ds (TweetRegressionDatasetFactory): the dataset used to crossvalidate the network.
:param iterations (int, optional): number of iterations for the crossvalidation.
:returns error (float): the average percent error of the dataset, tested on the network using crossvalidation.
"""
x = ds.DS['input']
y = ds.DS['target']
n, m = x.shape
errors = np.zeros(iterations)
cv = cross_validation.KFold(n, iterations, shuffle=True)
i = 0
for train_index, test_index in cv:
x_train = x[train_index, :]
y_train = y[train_index, :]
x_test = x[test_index, :]
y_test = y[test_index, :]
ds_train = RegressionData.conv2DS(x_train, y_train)
ds_test = RegressionData.conv2DS(x_test, y_test)
trainer = BackpropTrainer(
self.network,
dataset=ds_train,
learningrate=self.learningrate,
momentum=self.momentum,
verbose=self.verbose,
batchlearning=self.batchlearning)
trainer.trainUntilConvergence(
dataset=ds_train,
maxEpochs=self.max_epochs,
continueEpochs=self.con_epochs)
tstresult = self.test(ds_test)
errors[i] = tstresult[0]
i += 1
print "Simple Regression Neural Network cross-validation test errors: " % errors
return np.average(errors)
# Karol Dzitkowski # [email protected] # 10-10-2014 from pylab import * from csv_data import RegressionData from regression import RegressionNeuralNetwork from pybrain.utilities import percentError d_train = RegressionData('data/regress_1_train.csv') nn = RegressionNeuralNetwork(1, 1, 3, verbose=True, learningrate=0.025) nn.apply_custom_network([3,3]) #RUN NETWORK FOR CLASSIFICATION t = nn.run(d_train) print 'train error', nn.test(d_train) d_test = RegressionData('data/regress_1_tst.csv') print 'test error', nn.test(d_test) # get error # result = t.testOnClassData(dataset=d_train.DS) # error = percentError(result, d_train.DS['class']) # print 'error =', error # PLOT RESULTS figure(1) d_train.plotResult(nn) figure(2) d_test.plotResult(nn)