def get_hcond(self):
return kde_condentropy(self.input, K.exp(self.noiselayer.logvar))
def get_h(self):
totalvar = K.exp(self.noiselayer.logvar)+K.exp(self.kdelayer.logvar)
return kde_entropy(self.input, totalvar)
def logsumexp(mx, axis):
cmax = K.max(mx, axis=axis)
cmax2 = K.expand_dims(cmax, 1)
mx2 = mx - cmax2
return cmax + K.log(K.sum(K.exp(mx2), axis=1))
def on_train_begin(self, logs={}):
self.nlayerinput = lambda x: K.function([self.model.layers[0].input], [self.kdelayer.input])([x])[0]
N, dims = self.entropy_train_data.shape
Kdists = K.placeholder(ndim=2)
Klogvar = K.placeholder(ndim=0)
def obj(logvar, dists):
#print 'here', logvar # lossfunc([dists, logvar[0]])[0]
return lossfunc([dists, logvar.flat[0]])[0]
def jac(logvar, dists):
#print logvar, lossfunc([dists, logvar[0]]), jacfunc([dists, logvar[0]])
return np.atleast_2d(np.array(jacfunc([dists, logvar.flat[0]])))[0]
lossfunc = K.function([Kdists, Klogvar,], [kde_entropy_from_dists_loo(Kdists, N, dims, K.exp(Klogvar))])
jacfunc = K.function([Kdists, Klogvar,], K.gradients(kde_entropy_from_dists_loo(Kdists, N, dims, K.exp(Klogvar)), Klogvar))
self.obj =obj # lambda logvar, dists: np.array([lossfunc([dists, logvar[0]]),]) # [0]
self.jac =jac # lambda logvar, dists: jacfunc([dists, np.array([logvar]).flat[0]])[0]
def get_noise(self, x):
#if not hasattr(self, 'saved_noise'):
# self.saved_noise = K.random_normal(shape=K.shape(x), mean=0., std=1)
return K.exp(0.5*self.logvar) * K.random_normal(shape=K.shape(x), mean=0., std=1)