def KL(rho, rho_target, KL_flat):
y = rho.copy()
if KL_flat:
y[gp.where(y < rho_target)] = rho_target * gp.ones(
y[gp.where(y < rho_target)].shape)
return rho_target * gp.log(rho_target / y) + (1 - rho_target) * gp.log(
(1 - rho_target) / (1 - y))
def where(x, *args):
"""Delegate to gnumpy.where or numpy.where depending on the type of `x`."""
if not isinstance(x, np.ndarray):
return gp.where(x, *args)
else:
return np.where(x, *args)
def where(x, *args):
"""Delegate to gnumpy.where or numpy.where depending on the type of `x`."""
if not isinstance(x, np.ndarray):
return gp.where(x, *args)
else:
return np.where(x, *args)
def predict(self, X):
return np.where(self.net_input(X) >= 0.0, 1, -1)
def predict(self, X):
return np.where(self.activation(X) >= 0.0, 1, -1)
errors += int(update != 0.0)
self.errors_.append(errors)
return self
def net_input(self, X):
return np.dot(X, self.w_[1:]) + self.w_[0]
def predict(self, X):
return np.where(self.net_input(X) >= 0.0, 1, -1)
import pandas as pd
df = pd.read_csv('https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data', header=None)
# setosa and versicolor
y = df.iloc[0:100, 4].values
y = np.where(y == 'Iris-setosa', -1, 1)
# sepal length and petal length
X = df.iloc[0:100, [0,2]].values
import matplotlib.pyplot as plt
from mlxtend.evaluate import plot_decision_regions
ppn = Perceptron(epochs=10, eta=0.1)
ppn.train(X, y)
print('Weights: %s' % ppn.w_)
plot_decision_regions(X, y, clf=ppn)
plt.title('Perceptron')
plt.xlabel('sepal length [cm]')
def where(x):
check_type(x)
if is_np(x):
return np.where(x)
else:
return gp.where(x)
def where(x):
check_type(x)
if is_np(x):
return np.where(x)
else:
return gp.where(x)
def d_KL(rho, rho_target, KL_flat):
y = rho.copy()
if KL_flat:
y[gp.where(y < rho_target)] = rho_target * gp.ones(
y[gp.where(y < rho_target)].shape)
return -rho_target / y + (1 - rho_target) / (1 - y)
def relu_prime_truncated(x):
y = gp.ones(x.shape)
#if y>.9999: print 'salam'
y[gp.where(x > .9999)] = gp.zeros(x[gp.where(x > .9999)].shape)
y[gp.where(x < .00001)] = gp.zeros(x[gp.where(x < .00001)].shape)
return y
def relu_truncated(x):
y = x.copy()
y[gp.where(x > .9999)] = .9999 * gp.ones(x[gp.where(x > .9999)].shape)
y[gp.where(
x < .00001)] = .00001 * gp.ones(x[gp.where(x < .00001)].shape)
return y
