SiameseNetwork実行時のGPUの使用率

Question

kerasとtensorflowで実装したSiamese Networkを実行するとGPUの使用率が0～80％くらいを推移し、CPUの使用率が93％くらいになります。この値は正常なのでしょうか？
一応精度はそれぞれaccuracyが0.96、lossが0.03くらいは出ます。
GPUが使用できているかどうかと使えていない場合の原因が知りたいです。

ubuntu:18.04
GPU:GeForce GTX 1070 8GB
nvidia driver:495
cuda:10.0
cudnn:7.4
python:3.6
tensorflow:1.14.0
keras:2.2.5
tf.keras:2.2.4

import numpy as np
import random
import tensorflow as tf
import keras
from tensorflow.keras.datasets import mnist
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Input, Flatten, Dense, Dropout, Lambda
from tensorflow.keras.optimizers import RMSprop
from tensorflow.keras import backend as K
from time import time

t_start = time() # 開始時間

num_classes = 10 # 0～9
epochs = 20

def euclidean_distance(vects):
    x, y = vects
    sum_square = K.sum(K.square(x - y), axis=1, keepdims=True)
    return K.sqrt(K.maximum(sum_square, K.epsilon()))

def eucl_dist_output_shape(shapes):
    shape1, shape2 = shapes
    return (shape1[0], 1)

def contrastive_loss(y_true, y_pred):
    margin = 1
    square_pred = K.square(y_pred)
    margin_square = K.square(K.maximum(margin - y_pred, 0))
    return K.mean(y_true * square_pred + (1 - y_true) * margin_square)

def create_pairs(x, digit_indices):
    '''Positive and negative pair creation.
    Alternates between positive and negative pairs.
    '''
    pairs = []
    labels = []
    n = min([len(digit_indices[d]) for d in range(num_classes)]) - 1
    for d in range(num_classes):
        for i in range(n):
            z1, z2 = digit_indices[d][i], digit_indices[d][i + 1]
            pairs += [[x[z1], x[z2]]]
            inc = random.randrange(1, num_classes)
            dn = (d + inc) % num_classes
            z1, z2 = digit_indices[d][i], digit_indices[dn][i]
            pairs += [[x[z1], x[z2]]]
            labels += [1, 0]
    return np.array(pairs), np.array(labels)


def create_base_network(input_shape):
    '''Base network to be shared (eq. to feature extraction).
    '''
    input = Input(shape=input_shape)
    x = Flatten()(input)
    x = Dense(128, activation='relu')(x)
    x = Dropout(0.1)(x)
    x = Dense(128, activation='relu')(x)
    x = Dropout(0.1)(x)
    x = Dense(128, activation='relu')(x)
    return Model(input, x)

def compute_accuracy(y_true, y_pred):
    '''Compute classification accuracy with a fixed threshold on distances.
    '''
    pred = y_pred.ravel() < 0.5
    return np.mean(pred == y_true)


def accuracy(y_true, y_pred):
    '''Compute classification accuracy with a fixed threshold on distances.
    '''
    return K.mean(K.equal(y_true, K.cast(y_pred < 0.5, y_true.dtype)))

# The data, split between train and test sets
mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train = x_train / 255.0
x_test = x_test / 255.0
# input_shape = (28, 28, 1)
input_shape = x_train.shape[1:] # (28, 28)

# create training+test positive and negative pairs
digit_indices = [np.where(y_train == i)[0] for i in range(num_classes)]
tr_pairs, tr_y = create_pairs(x_train, digit_indices)

digit_indices = [np.where(y_test == i)[0] for i in range(num_classes)]
te_pairs, te_y = create_pairs(x_test, digit_indices)

print(np.shape(tr_pairs))

base_network = create_base_network(input_shape)

input_a = Input(shape=input_shape)
input_b = Input(shape=input_shape)

# because we re-use the same instance `base_network`,
# the weights of the network
# will be shared across the two branches
processed_a = base_network(input_a)
processed_b = base_network(input_b)

distance = Lambda(euclidean_distance,
                  output_shape=eucl_dist_output_shape)([processed_a, processed_b])

model = Model([input_a, input_b], distance)

model.summary()
tf.keras.utils.plot_model(model, to_file = 'model.png', show_shapes = True, show_layer_names = True)

# train
tr_y = tf.cast(tr_y, dtype='float32')
te_y = tf.cast(te_y, dtype='float32')

rms = RMSprop()
model.compile(loss=contrastive_loss, optimizer=rms, metrics=[accuracy])

H = model.fit([tr_pairs[:, 0], tr_pairs[:, 1]], tr_y,
          batch_size=128,
          epochs=epochs,
          steps_per_epoch=1000
          )

t_end = time() #終了時間
t_elapsed = t_end - t_start

print("処理時間は{0}".format(t_elapsed))

Comments

[ppaul]

どういう答えを期待しているのでしょう？
「はい／いいえ」ですか。

Answer 1

GPUを無効化して演算時間を比較するが手っ取り早いと思います。

GPUもフルタイム全力で回っているわけではないので、使用率が上下するのはよくあることかな？とも思います。