def __init__(self,
x,
y,
x0,
h_mean,
h_cov,
h_cov_dia,
fve=0.85,
binning=True,
bin_weight=True,
ker_fun='Epan',
bw_select='Partition',
dtype='f4'):
self.__Check_Input(x, y, x0, h_mean, h_cov, h_cov_dia, fve, binning,
bin_weight, ker_fun, bw_select, dtype)
self.__num_fun = len(x)
self.__d = x0.shape[-1]
self.__grid_shape = np.asarray(x0.shape[:-1])
self.__bin_width = np.ptp(x0.reshape(-1, self.__d),
0) / (self.__grid_shape - 1)
self.__n_train = lpr.Partition_Data_Size(len(x))
x0_min = x0.reshape(-1, self.__d).min(0)
x0_max = x0.reshape(-1, self.__d).max(0)
self.__grid = tuple(
np.linspace(x0_min.take(i), x0_max.take(i), x0.shape[i])
for i in range(self.__d))
self.__Main(x, y, x0, h_mean, h_cov, h_cov_dia, fve, binning,
bin_weight, ker_fun, bw_select, dtype)
self.__Standardize_Output()
def __init__(self,
X_data,
h_mean,
h_cov,
binning=True,
bin_weight=True,
ker_fun='Gaussian'):
num_grid = X_data.time_grid.shape[0]
tran_size = X_data.obs_tran.shape[0]
num_pts = X_data.tran_time_pts.shape[1]
x = X_data.tran_time_pts.reshape(tran_size, num_pts, 1)
y = X_data.obs_tran
x0 = X_data.time_grid.reshape(num_grid, 1)
self.__Check_Input(x, y, x0, h_mean, h_cov, binning, bin_weight,
ker_fun)
self.__num_fun, self.__num_pt, self.__d = tran_size, num_pts, 1
self.__grid_shape = np.asarray(x0.shape[:-1])
self.__bin_width = np.ptp(x0.reshape(-1, self.__d),
0) / (self.__grid_shape - 1)
self.__n_train = lpr.Partition_Data_Size(self.__num_fun *
self.__num_pt -
np.isnan(y).sum())
x0_min = x0.reshape(-1, self.__d).min(0)
x0_max = x0.reshape(-1, self.__d).max(0)
self.__grid = tuple(
np.linspace(x0_min.take(i), x0_max.take(i), x0.shape[i])
for i in range(self.__d))
self.__Main(x, y, x0, h_mean, h_cov, binning, bin_weight, ker_fun)
self.__Standardize_Output()
def __CV_Cov_Partition(self, x, y, x0, h, ker_fun):
grid_shape, d = np.asarray(x0.shape[:-1]), x0.shape[-1]
n_grid = np.prod(self.__grid_shape)
x_displacement = (
(x - x0.reshape(-1, d).min(axis=0)) / self.__bin_width +
np.ones(self.__d) / 2).astype(np.int32)
x_p = np.sum(x_displacement *
np.append(grid_shape[::-1].cumprod()[-2::-1], 1),
axis=2)
xx_p = (x_p.repeat(x_p.shape[1], 1) * n_grid +
np.tile(x_p, x_p.shape[1]))
xx_p = np.delete(xx_p, np.arange(0, self.__num_pt**2,
self.__num_pt + 1), 1)
yy = np.einsum('ij,ik->ijk', y, y).reshape(self.__num_fun, -1)
yy = np.delete(yy, np.arange(0, self.__num_pt**2, self.__num_pt + 1),
1)
non_nan_value = ~np.isnan(yy).reshape(-1)
n_real_val = non_nan_value.sum()
xx_p = np.compress(non_nan_value, xx_p.reshape(-1))
yy = np.compress(non_nan_value, yy.reshape(-1))
n_train = lpr.Partition_Data_Size(n_real_val)
random_order = np.random.permutation(n_real_val)
train_xx, test_xx = np.split(
xx_p.reshape(-1).take(random_order), [n_train])
train_yy, test_yy = np.split(
yy.reshape(-1).take(random_order), [n_train])
bin_xx = np.bincount(train_xx, minlength=n_grid**2)
bin_yy = np.bincount(train_xx, train_yy, minlength=n_grid**2)
bin_xx = bin_xx.reshape(np.tile(self.__grid_shape, 2))
bin_yy = bin_yy.reshape(np.tile(self.__grid_shape, 2))
ssq = np.zeros(h.shape[0])
for i in range(h.shape[0]):
fit_y = lpr.Lpr_For_Bin([bin_yy, bin_xx],
np.tile(self.__bin_width, 2),
h.take(i, 0),
ker_fun=ker_fun)
ssq[i] = ((test_yy - fit_y[test_xx])**2).sum()
if np.isnan(fit_y).any():
ssq[i] = np.nan
h_opt = h.take(np.nanargmin(ssq), 0)
return ([h_opt, xx_p, yy])