def invert(self,
data,
input=None,
mask=None,
tag=None,
clip=np.array([0.0, 1.0])):
# yhat = self.link(np.clip(data, .1, np.inf))
if self.bin_size < 1:
yhat = smooth(data, 20)
xhat = self.link(np.clip(yhat, 0.01, np.inf))
xhat = self._invert(xhat, input=input, mask=mask, tag=tag)
xhat = smooth(xhat, 10)
if self.bin_size < 1:
# xhat = np.clip(xhat, -0.9, 0.9)
xhat = np.clip(xhat, -1.02, 1.02)
# for t in range(xhat.shape[0]):
# if np.all(xhat[np.max([0,t-2]):t+3]>0.99) and t>2:
# # if np.median(xhat[np.max([0,t-2]):t+3])>0.99 and t>0:
# xhat[np.minimum(0,t-2):] = 1.01*np.ones(np.shape(xhat[np.minimum(0,t-2):]))
# xhat[:np.maximum(0,t-2)] = 0.0*np.ones(np.shape(xhat[:np.maximum(0,t-2)]))
# break
# if np.all(xhat[np.max([0,t-2]):t+3]<-0.99) and t>2:
# # if np.median(xhat[np.max([0,t-2]):t+3])>0.99 and t>0:
# xhat[np.minimum(0,t-2):] = -1.01*np.ones(np.shape(xhat[np.minimum(0,t-2):]))
# xhat[:np.maximum(0,t-2)] = 0.0*np.ones(np.shape(xhat[:np.maximum(0,t-2)]))
# break
if np.abs(xhat[0]).any() > 1.0:
xhat[0] = 0.05 * npr.randn(1, self.D)
return xhat
def invert(self,
data,
input=None,
mask=None,
tag=None,
clip=np.array([0.0, 1.0])):
yhat = smooth(data, 20)
xhat = self.link(np.clip(yhat, 0.01, np.inf))
xhat = self._invert(xhat, input=input, mask=mask, tag=tag)
num_pad = 10
xhat = smooth(np.concatenate((np.zeros((num_pad, self.D)), xhat)),
10)[num_pad:, :]
xhat = np.clip(xhat, -1.1, 1.1)
if np.abs(xhat[0]).any() > 1.0:
xhat[0] = 0.1 * npr.randn(1, self.D)
return xhat
def invert(self, data, input=None, mask=None, tag=None):
yhat = smooth(data,20)
xhat = self.link(np.clip(yhat, 0.01, np.inf))
xhat = self._invert(xhat, input=input, mask=mask, tag=tag)
for t in range(xhat.shape[0]):
if np.all(xhat[np.max([0,t-2]):t+3]>0.99) and t>2:
xhat[np.minimum(0,t-2):] = 1.01*np.ones(np.shape(xhat[np.minimum(0,t-2):]))
xhat[:np.maximum(0,t-2)] = np.clip(xhat[:np.maximum(0,t-2)], -0.5,0.95)
if np.abs(xhat[0])>1.0:
xhat[0] = 0.5 + 0.01*npr.randn(1,np.shape(xhat)[1])
return xhat
def invert(self,
data,
input=None,
mask=None,
tag=None,
clip=np.array([0.0, 1.0])):
# yhat = self.link(np.clip(data, .1, np.inf))
if self.bin_size < 1:
yhat = smooth(data, 20)
xhat = self.link(np.clip(yhat, 0.01, np.inf))
xhat = self._invert(xhat, input=input, mask=mask, tag=tag)
xhat = smooth(xhat, 10)
if self.bin_size < 1:
xhat = np.clip(xhat, -0.95, 0.95)
# in all models, x starts in between boundaries at [-1,1]
if np.abs(xhat[0]).any() > 1.0:
xhat[0] = 0.05 * npr.randn(1, self.D)
return xhat