オートエンコーダー(AutoEncoder)について質問があり投稿しました。
初めての投稿で不備があったら教えて頂けると幸いです。

以下のプログラムを実装したのですが、今、横160縦120pixの画像を入力すると、「ResourceExhaustedError」が発生して、学習に進むことができません。
具体的には、130行目のところでErrorが発生します。
一方、解像度を半分の横80縦60pixにすると、EPOCが進み学習が進んでいるように見えます。
(プログラムで画像を割る2して小さくしています。)

画像サイズ(横160縦120pix)や枚数(約700枚)が特に多くないと思っているのですが、なぜエラーが発生するのかと解決方法をご教授頂けないでしょうか。
メインメモリ不足が影響している可能性を考えてメモリを128GBにしましたが同様のエラーが発生します。

以上、よろしくお願いいたします。

実装環境を以下に記します。
CPU:Xeon E5-1620v4 4core/8t
マザーボード:ASUS X99-E WS
メモリ:DDR4-2400 64GB(8G×8)
GPU:NVIDIA Quadro GP100 16GB 2個
OS:ubuntu16.04LTS

ソースコード

import numpy as np
import tensorflow as tf
from tensorflow.python.framework import ops
import cv2
import os

DATASET_PATH = "/home/densos/workspaces/autoencoder"
DIR_PATH = "input_gray_160*120"
IMAGE_PATH = os.path.join(DATASET_PATH, DIR_PATH)
X_PIXEL, Y_PIXEL = 160, 120
M = 1
N_HIDDENS = np.array(np.array([1.5]) * X_PIXEL * Y_PIXEL // (M*M), dtype = np.int)
TRANCE_FRAME_NUM = 700

ops.reset_default_graph()

def xavier_init(fan_in, fan_out, constant = 1):
    low = -constant * np.sqrt(6.0 / (fan_in + fan_out))
    high = constant * np.sqrt(6.0 / (fan_in + fan_out))
    return tf.random_uniform((fan_in, fan_out), minval = low, maxval = high, dtype = tf.float32)

class AdditiveGaussianNoiseAutoencoder(object):
    def __init__(self, n_input, n_hidden, transfer_function = tf.nn.sigmoid, optimizer = tf.train.AdamOptimizer(), scale = 0.1):
        self.n_input = n_input
        self.n_hidden = n_hidden
        self.transfer = transfer_function
        self.scale = tf.placeholder(tf.float32)
        self.training_scale = scale
        network_weights = self._initialize_weights()
        self.weights = network_weights
        self.sparsity_level = np.repeat([0.05], self.n_hidden).astype(np.float32)
        self.sparse_reg = 0.1

        # model
        self.x = tf.placeholder(tf.float32, [None, self.n_input])
        self.hidden = self.transfer(tf.add(tf.matmul(self.x + scale * tf.random_normal((n_input,)),
                self.weights['w1']),
                self.weights['b1']))
        self.reconstruction = tf.add(tf.matmul(self.hidden, self.weights['w2']), self.weights['b2'])

        # cost
        self.cost = 0.5 * tf.reduce_sum(tf.pow(tf.subtract(self.reconstruction, self.x), 2.0)) + self.sparse_reg \
                        * self.kl_divergence(self.sparsity_level, self.hidden)

        self.optimizer = optimizer.minimize(self.cost)

        init = tf.global_variables_initializer()
        self.sess = tf.Session()
        self.sess.run(init)

    def _initialize_weights(self):
        all_weights = dict()
        all_weights['w1'] = tf.Variable(xavier_init(self.n_input, self.n_hidden))
        all_weights['b1'] = tf.Variable(tf.zeros([self.n_hidden], dtype = tf.float32))
        all_weights['w2'] = tf.Variable(tf.zeros([self.n_hidden, self.n_input], dtype = tf.float32))
        all_weights['b2'] = tf.Variable(tf.zeros([self.n_input], dtype = tf.float32))
        return all_weights

    def partial_fit(self, X):
        cost, opt = self.sess.run((self.cost, self.optimizer), feed_dict = {self.x: X,
                                                                            self.scale: self.training_scale
                                                                            })
        return cost

    def kl_divergence(self, p, p_hat):
        return tf.reduce_mean(p * tf.log(p) - p * tf.log(p_hat) + (1 - p) * tf.log(1 - p) - (1 - p) * tf.log(1 - p_hat))

    def calc_total_cost(self, X):
        return self.sess.run(self.cost, feed_dict = {self.x: X,
                                                     self.scale: self.training_scale
                                                     })

    def transform(self, X):
        return self.sess.run(self.hidden, feed_dict = {self.x: X,
                                                       self.scale: self.training_scale
                                                       })

    def generate(self, hidden = None):
        if hidden is None:
            hidden = np.random.normal(size = self.weights["b1"])
        return self.sess.run(self.reconstruction, feed_dict = {self.hidden: hidden})

    def reconstruct(self, X):
        return self.sess.run(self.reconstruction, feed_dict = {self.x: X,
                                                               self.scale: self.training_scale
                                                               })

    def getWeights(self):
        return self.sess.run(self.weights['w1'])

    def getBiases(self):
        return self.sess.run(self.weights['b1'])

def get_random_block_from_data(data, batch_size):
    start_index = np.random.randint(0, len(data) - batch_size)
    return data[start_index:(start_index + batch_size)]


if __name__ == '__main__':
#get input data lists
    lists = []
    for file in os.listdir(IMAGE_PATH):
        if file.endswith(".jpeg"):
            lists.append(file)
        lists.sort()

#read input data    
    input_images = []
    for image in lists:
        tmp = cv2.imread(os.path.join(IMAGE_PATH, image), cv2.IMREAD_GRAYSCALE)
        tmp = cv2.resize(tmp, (X_PIXEL // M, Y_PIXEL // M))
        tmp = tmp.reshape(tmp.shape[0] * tmp.shape[1])
        input_images.append(tmp)

#preprocess images    
    input_images = np.array(input_images) / 255.

#convert data to float16
    input_images = np.array(input_images, dtype = np.float16)

#set train and test data
    X_train = input_images[:500]
    X_test = input_images[500:]

    n_samples = X_train.shape[0]
    training_epochs = 200
    batch_size = X_train.shape[0] // 4
    display_step = 10

    autoencoder = AdditiveGaussianNoiseAutoencoder(n_input = X_train.shape[1],
                                                   n_hidden = N_HIDDENS[0],
                                                   transfer_function = tf.nn.relu6,
                                                   optimizer = tf.train.AdamOptimizer(learning_rate = 0.001),
                                                   scale = 0.01)

    for epoch in range(training_epochs):
        avg_cost = 0.
        total_batch = int(n_samples / batch_size)
        # Loop over all batches
        for i in range(total_batch):
            batch_xs = get_random_block_from_data(X_train, batch_size)

            # Fit training using batch data
            cost = autoencoder.partial_fit(X_train)
            # Compute average loss
            avg_cost += cost / n_samples * batch_size

        # Display logs per epoch step
        if epoch % display_step == 0:
            print("Epoch:", '%04d' % (epoch + 1), "cost=", avg_cost)

    print("Finish Train")

predicted_imgs = autoencoder.reconstruct(X_test)
predicted_imgs = np.array((predicted_imgs) * 255, dtype = np.uint8)
input_imgs = np.array((X_test) * 255, dtype = np.uint8)

# plot the reconstructed images
for i in range(100):
    im1 = predicted_imgs[i].reshape((Y_PIXEL//M, X_PIXEL//M))
    im2 = input_imgs[i].reshape((Y_PIXEL//M, X_PIXEL//M))

    img_v_union = cv2.vconcat([im1, im2])
    cv2.moveWindow('result.jpg', 100, 200)
    cv2.imshow('result.jpg', img_v_union)

    cv2.waitKey(33)