Created on May 22, 2017
@author: a3438
'''
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from SimulateModelData import simulateModelData
from sklearn.preprocessing import StandardScaler
import sklearn.neural_network
if __name__ == '__main__':
n_samples = 10
nFeatures = 2
numClasses = 2
nld = simulateModelData(nFeatures, n_samples)
nld.simulateClassification(numClasses)
ss = StandardScaler()
ss.fit(nld.X)
XNew = ss.transform(nld.X)
print nld.X
print nld.y
clf = sklearn.neural_network.MLPClassifier(solver='lbfgs',
alpha=1e-5,
hidden_layer_sizes=(5, 2),
random_state=1)
clf.fit(XNew, nld.y)
y1 = clf.predict(XNew)
print np.max(y1 - nld.y), np.average(y1 - nld.y)
'''
Created on Mar 7, 2017
@author: a3438
'''
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from SimulateModelData import simulateModelData
from sklearn.svm import SVR
from sklearn.preprocessing import StandardScaler
if __name__ == '__main__':
n_samples = 200
n_features = 2
nld = simulateModelData(n_features, n_samples)
nld.simulateNonLinear(sigma=0.2)
ss = StandardScaler()
ss.fit(nld.X)
XNew = ss.transform(nld.X)
clf = SVR(C=1.0, epsilon=0.1)
clf.fit(XNew, nld.y)
y1 = clf.predict(XNew)
print np.max(y1 - nld.y), np.average(y1 - nld.y)
nld2 = simulateModelData(n_features, n_samples)
nld2.simulateNonLinear(sigma=0.2)
XNew2 = ss.transform(nld2.X)
y2 = clf.predict(XNew2)
print np.max(y2 - nld2.y), np.average(y2 - nld2.y)
np.random.seed(0)
# Generate datasets. We choose the size big enough to see the scalability
# of the algorithms, but not too big to avoid too long running times
n_samples = 1500
n_features = 10
n_classes = 4
noisy_circles = datasets.make_circles(n_samples=n_samples,
factor=.5,
noise=.05)
noisy_moons = datasets.make_moons(n_samples=n_samples, noise=.05)
blobs = datasets.make_blobs(n_samples=n_samples, random_state=8)
meanVectors = np.random.uniform(0.0, 10.0, size=[n_classes, n_features])
sigmaVectors = 4.0 * np.ones([n_classes, n_features])
cd = simulateModelData(n_features, n_samples)
cd.simulateClass(n_classes, mu=meanVectors, sigma=sigmaVectors)
normalClusters = (cd.X, cd.y)
no_structure = np.random.rand(n_samples, 2), None
colors = np.array([x for x in 'bgrcmykbgrcmykbgrcmykbgrcmyk'])
#colors = np.hstack([colors] * 20)
datasets = [noisy_circles, noisy_moons, blobs, normalClusters]
plt.figure(figsize=(len(datasets), 9.5))
plt.subplots_adjust(left=.02,
right=.98,
bottom=.001,
top=.96,