RIM guassians 2

gaussians2.py

# Create training, validation, and test sets
train_percentage = 0.7
valid_percentage = 0.9
test_percentage = 1.0
len_X = len(gaussians_initial)
# Training
X_train = gaussians_initial[:int(train_percentage*len_X)]
Y_train = gaussians_final[:int(train_percentage*len_X)]
A_train = powerlaw_conv[:int(train_percentage*len_X)]
N_train = [np.diag(noise_val) for noise_val in noise[:int(train_percentage*len_X)]]
#Validation
X_valid = gaussians_initial[int(train_percentage*len_X):int(valid_percentage*len_X)]
Y_valid = gaussians_final[int(train_percentage*len_X):int(valid_percentage*len_X)]
A_valid = powerlaw_conv[int(train_percentage*len_X):int(valid_percentage*len_X)]
N_valid = [np.diag(noise_val) for noise_val in noise[int(train_percentage*len_X):int(valid_percentage*len_X)]]
#Test
X_test = gaussians_initial[int(valid_percentage*len_X):]
Y_test = gaussians_final[int(valid_percentage*len_X):]
A_test = powerlaw_conv[int(valid_percentage*len_X):]
N_test = [np.diag(noise_val) for noise_val in noise[int(valid_percentage*len_X):]]