def Linear(x):
return tf.layers.dense(inputs=x, units=dt.getsinifcount(), name='linear')
x = flatten(x)
x = Linear(x)
# x = tf.reshape(x, [-1, 10])
return x
if __name__ == '__main__':
dt = dt.Data()
train_x, train_y = dt.read_train_images(120, 120)
test_x, test_y = dt.read_test_images(120, 120)
#train_x, test_x = color_preprocessing(train_x, test_x)
# image_size = 32, img_channels = 3, class_num = 10 in cifar10
x = tf.placeholder(tf.float32, shape=[None, 120, 120, 3])
label = tf.placeholder(tf.float32, shape=[None, dt.getsinifcount()])
training_flag = tf.placeholder(tf.bool)
learning_rate = tf.placeholder(tf.float32, name='learning_rate')
logits = DenseNet(x=x,
nb_blocks=nb_block,
filters=growth_k,
training=training_flag).model
cost = tf.reduce_mean(
tf.nn.softmax_cross_entropy_with_logits_v2(labels=label,
logits=logits))
"""
l2_loss = tf.add_n([tf.nn.l2_loss(var) for var in tf.trainable_variables()])
optimizer = tf.train.MomentumOptimizer(learning_rate=learning_rate, momentum=nesterov_momentum, use_nesterov=True)