import models
import numpy as np
np.random.seed(123)
print("Loading data . . . ")
(_, Y_train), (_, Y_test) = models.load_data_from_keras()
X_train = np.load("results/vgg16_ae_X_train.npy")
X_test = np.load("results/vgg16_ae_X_test.npy")
"""
n = 1000
X_train = X_train[:n]
Y_train = Y_train[:n]
X_test = X_test[:n]
Y_test = Y_test[:n]
"""
"""
m = X_train.min()
s = X_train.max()
X_train -= m
X_train = X_train/s
X_test -= m
X_test = X_test/s
"""
print("Training . . . ")
model = models.vgg16(lr=1e-3, mbs=50, pred_mbs=500, seed=456)
model.start_session()
accs = model.train(X_train,
Y_train,
eval_set=(X_test, Y_test),
import models
import numpy as np
print("Loading data . . . ")
#(X_train, Y_train), (X_test, Y_test) = models.load_data()
(X_train, Y_train), _ = models.load_data_from_keras()
print("Resizing . . . ")
X_train = models.preprocess.resize(X_train)
print("Getting center . . . ")
center = np.mean(X_train, axis=(0, 1, 2))
print("Writing results . . . ")
np.savetxt("results/center_after_resize.txt", center)
print("Done")