从Keras的生成器获取x_test，y_test?

问题描述

对于某些问题，验证数据不能是生成器，例如: TensorBoard直方图:

For certain problems, the validation data can't be a generator, e.g.: TensorBoard histograms:

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, ..)

---

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

---

How do I provide it as validation_data=(x_test, y_test)?

推荐答案

Python 2.7和Python 3. *解决方案:

Python 2.7 and Python 3.* solution:

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'，然后:

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/7f6cc03ed6c3dd84548a039e2e0>

Full runnable code: https://gist.github.com/AlecTaylor/7f6cc03ed6c3dd84548a039e2e0fd006

这篇关于从Keras的生成器获取x_test，y_test?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！