从 Keras 中的生成器获取 x_test、y_test?

Getting x_test, y_test from generator in Keras?

对于某些问题,验证数据不能是生成器,例如:TensorBoard histograms:

If printing histograms, validation_data must be provided, and cannot be a generator.

我当前的代码如下:

image_data_generator = ImageDataGenerator()

training_seq   = image_data_generator.flow_from_directory(training_dir)
validation_seq = image_data_generator.flow_from_directory(validation_dir)
testing_seq    = image_data_generator.flow_from_directory(testing_dir)

model = Sequential(..)
# ..
model.compile(..)
model.fit_generator(training_seq, validation_data=validation_seq, ..)

如何将其提供为 validation_data=(x_test, y_test)

Python 2.7 和 Python 3.* 解决方案:

from platform import python_version_tuple

if python_version_tuple()[0] == '3':
    xrange = range
    izip = zip
    imap = map
else:
    from itertools import izip, imap

import numpy as np

# ..
# other code as in question
# ..

x, y = izip(*(validation_seq[i] for i in xrange(len(validation_seq))))
x_val, y_val = np.vstack(x), np.vstack(y)

或支持class_mode='binary',则:

from keras.utils import to_categorical

x_val = np.vstack(x)
y_val = np.vstack(imap(to_categorical, y))[:,0] if class_mode == 'binary' else y

完整的可运行代码:https://gist.github.com/AlecTaylor/7f6cc03ed6c3dd84548a039e2e0fd006

更新(2018 年 6 月 22 日):阅读 OP 提供的答案以获得简洁高效的解决方案。阅读我的以了解发生了什么。


在 python 中,您可以使用以下方法获取所有发电机数据:

data = [x for x in generator]

但是,ImageDataGenerators 不会终止,因此上面的方法不起作用。但是我们可以使用相同的方法并进行一些修改以在这种情况下工作:

data = []     # store all the generated data batches
labels = []   # store all the generated label batches
max_iter = 100  # maximum number of iterations, in each iteration one batch is generated; the proper value depends on batch size and size of whole data
i = 0
for d, l in validation_generator:
    data.append(d)
    labels.append(l)
    i += 1
    if i == max_iter:
        break

现在我们有两个张量批次列表。我们需要重塑它们以制作两个张量,一个用于数据(即 X),一个用于标签(即 y):

data = np.array(data)
data = np.reshape(data, (data.shape[0]*data.shape[1],) + data.shape[2:])

labels = np.array(labels)
labels = np.reshape(labels, (labels.shape[0]*labels.shape[1],) + labels.shape[2:])