我试图循环训练1000倍的顺序模型。在每个循环中,我的程序都会泄漏内存,直到用尽并收到OOM异常。

我之前已经问过类似的问题
(Training multiple Sequential models in a row slows down)

并看到其他人也遇到类似的问题(Keras: Out of memory when doing hyper parameter grid search)

解决方案始终是在使用完模型后将K.clear_session()添加到您的代码中。所以我在上一个问题中做到了,但我仍在泄漏内存

这是重现此问题的代码。

import random
import time
from keras.models import Sequential
from keras.layers import Dense
from keras import backend as K
import tracemalloc


def run():
    tracemalloc.start()
    num_input_nodes = 12
    num_hidden_nodes = 8
    num_output_nodes = 1

    random_numbers = random.sample(range(1000), 50)
    train_x, train_y = create_training_dataset(random_numbers, num_input_nodes)

    for i in range(100):
        snapshot = tracemalloc.take_snapshot()
        for j in range(10):
            start_time = time.time()
            nn = Sequential()
            nn.add(Dense(num_hidden_nodes, input_dim=num_input_nodes, activation='relu'))
            nn.add(Dense(num_output_nodes))
            nn.compile(loss='mean_squared_error', optimizer='adam')
            nn.fit(train_x, train_y, nb_epoch=300, batch_size=2, verbose=0)
            K.clear_session()
            print("Iteration {iter}. Current time {t}. Took {elapsed} seconds".
                  format(iter=i*10 + j + 1, t=time.strftime('%H:%M:%S'), elapsed=int(time.time() - start_time)))

        top_stats = tracemalloc.take_snapshot().compare_to(snapshot, 'lineno')

        print("[ Top 5 differences ]")
        for stat in top_stats[:5]:
            print(stat)


def create_training_dataset(dataset, input_nodes):
    """
    Outputs a training dataset (train_x, train_y) as numpy arrays.
    Each item in train_x has 'input_nodes' number of items while train_y items are of size 1
    :param dataset: list of ints
    :param input_nodes:
    :return: (numpy array, numpy array), train_x, train_y
    """
    data_x, data_y = [], []
    for i in range(len(dataset) - input_nodes - 1):
        a = dataset[i:(i + input_nodes)]
        data_x.append(a)
        data_y.append(dataset[i + input_nodes])
    return numpy.array(data_x), numpy.array(data_y)

run()

这是我从第一个内存调试打印中获得的输出

/tensorflow/python/framework/ops.py:121:大小= 3485 KiB(+3485 KiB),计数= 42343(+42343)
/tensorflow/python/framework/ops.py:1400:size = 998 KiB(+998 KiB),count = 8413(+8413)
/tensorflow/python/framework/ops.py:116:size = 888 KiB(+888 KiB),count = 32468(+32468)
/tensorflow/python/framework/ops.py:1185:size = 795 KiB(+795 KiB),count = 3179(+3179)
/tensorflow/python/framework/ops.py:2354:size = 599 KiB(+599 KiB),count = 5886(+5886)

系统信息:
  • python 3.5
  • keras(1.2.2)
  • tensorflow (1.0.0)
  • 最佳答案

    内存泄漏源于Keras和TensorFlow,它们使用单​​个“默认图”来存储网络结构,随着内部for循环的每次迭代,网络结构的大小都会增加。

    调用K.clear_session()会释放两次与两次迭代之间的默认图相关联的(后端)状态,但是需要额外调用 tf.reset_default_graph() 才能清除Python状态。

    请注意,可能有一个更有效的解决方案:由于nn不依赖于任何一个循环变量,因此您可以在循环外部定义它,并在循环内部重用相同的实例。如果执行此操作,则无需清除 session 或重置默认图形,并且性能会提高,因为您可以受益于两次迭代之间的缓存。

    关于python-3.x - Keras(TensorFlow,CPU): Training Sequential models in loop eats memory,我们在Stack Overflow上找到一个类似的问题:https://stackoverflow.com/questions/42886049/

    10-12 19:29