Question

我尝试过转学，然后我得到了＃34;没有属性＆＃39; image_data_generator＆＃39;＆＃34;错误。这个错误意味着什么？

您可以通过以下方式重现我的错误。

*重现步骤

$ cd /tmp
$ git clone --branch no-attribute-error https://github.com/y-zono/btc-bch-ml.git
$ cd btc-bch-ml
$ docker build -t model-builder .
$ docker run -v /tmp/btc-bch-ml/docker:/docker -d -it model-builder
$ docker exec -it CONTAINER_ID /bin/bash
# cd /docker
# python3 ./train-predict.py

Traceback (most recent call last):
  File "./train-predict.py", line 62, in <module>
    history = model.fit_generator(img_itr_train, steps_per_epoch=steps_per_epoch, epochs=n_epoch)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/keras/_impl/keras/engine/training.py", line 1598, in fit_generator
    initial_epoch=initial_epoch)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/keras/_impl/keras/engine/training_generator.py", line 162, in fit_generator
    generator_output = next(output_generator)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/keras/_impl/keras/utils/data_utils.py", line 613, in get
    six.raise_from(StopIteration(e), e)
  File "<string>", line 3, in raise_from
StopIteration: 'ImageDataGenerator' object has no attribute 'image_data_generator'

*注

Tensorflow版本是＆＃34; tensorflow-1.8.0＆＃34;

我将所有代码和培训数据发布到github中。

https://github.com/y-zono/btc-bch-ml/blob/no-attribute-error/docker/train-predict.py#L62

Answer 1

当使用tensorflow1.8时，出现了同样的错误。如果删除传递给ImageDatagenerator的预处理功能，则您的代码可以使用，但是，您必须以其他方式对图像进行预处理。

Answer 2

更改

idg_train = ImageDataGenerator(rescale=1/255., 
    shear_range=0.1,
    zoom_range=0.1,
    horizontal_flip=True,
    preprocessing_function=preprocess_input)

到

idg_train = ImageDataGenerator(rescale=1/255., 
    shear_range=0.1, 
    zoom_range=0.1, 
    horizontal_flip=True)

Answer 3

我想我已经解决了你的问题。由于其中一个import语句，因此发生此问题。我建议你更换

from tensorflow.python.keras.applications.resnet50 import preprocess_input

到

from keras.preprocessing.image import ImageDataGenerator

这在我尝试的时候有效。如果你还面临问题，请告诉我。

Answer 4

我在tensorflow-1.8.0中遇到了相同的问题。最快的方法是创建一个包含您的预处理功能的空类，并将其传递给ImageDataGenerator实例的image_data_generator对象：

def preprocess_input(im_array):
    return im_array/255-.5

class foo(object):
    pass

imgen = ImageDataGenerator(preprocessing_function='pass anything')

imgen.image_data_generator = foo
imgen.image_data_generator.preprocessing_function = preprocess_input

flowgen = imgen.flow_from_directory('/tmp/training_folder/') 

flowgen.__getitem__(0)[0] #check result

另一种解决方法是通过删除tensorflow/python/keras/_impl/keras/preprocessing/image.py第672行来修复image_data_generator中的源代码

＆＃39; ImageDataGenerator＆＃39;对象没有属性＆＃39; image_data_generator＆＃39;

4 个答案: