keras的ImageDataGenerator和flow()的用法说明

ゝ一世哀愁。 2022-12-26 00:58 101阅读 0赞

更多编程教程请到：[菜鸟教程][Link 1] https://www.piaodoo.com/

友情链接：

[高州阳光论坛][Link 2]https://www.hnthzk.com/

[人人影视][Link 3]http://www.op-kg.com/

**ImageDataGenerator的参数自己看文档**

from keras.preprocessing import image
    import numpy as np

X\_train=np.ones((3,123,123,1))  
Y\_train=np.array(\[\[1\],\[2\],\[2\]\])  
generator=image.ImageDataGenerator(featurewise\_center=False,  
samplewise\_center=False,  
featurewise\_std\_normalization=False,  
samplewise\_std\_normalization=False,  
zca\_whitening=False,  
zca\_epsilon=1e-6,  
rotation\_range=180,  
width\_shift\_range=0.2,  
height\_shift\_range=0.2,  
shear\_range=0,  
zoom\_range=0.001,  
channel\_shift\_range=0,  
fill\_mode=‘nearest’,  
cval=0.,  
horizontal\_flip=True,  
vertical\_flip=True,  
rescale=None,  
preprocessing\_function=None,  
data\_format=‘channels\_last’)

a=generator.flow(X\_train,Y\_train,batch\_size=20)\#生成的是一个迭代器，可直接用于for循环  
‘’’  
batch\_size如果小于X的第一维m，next生成的多维矩阵的第一维是为batch\_size，输出是从输入中随机选取batch\_size个数据  
batch\_size如果大于X的第一维m，next生成的多维矩阵的第一维是m，输出是m个数据，不过顺序随机  
，输出的X，Y是一一对对应的  
如果要直接用于tf.placeholder()，要求生成的矩阵和要与tf.placeholder相匹配

‘’’  
X,Y=next(a)

print(Y)  
X,Y=next(a)

输出

[[2]
     [1]
     [2]]

\[\[2\]  
\[2\]  
\[1\]\]

**补充知识：****tensorflow 与keras 混用之坑**

在使用tensorflow与keras混用是model.save 是正常的但是在load\_model的时候报错了在这里mark 一下

其中错误为：TypeError: tuple indices must be integers, not list

再一一番百度后无结果，上谷歌后找到了类似的问题。但是是一对鸟文不知道什么东西（翻译后发现是俄文）。后来谷歌翻译了一下找到了解决方法。故将原始问题文章贴上来警示一下

原训练代码

from tensorflow.python.keras.preprocessing.image import ImageDataGenerator
    from tensorflow.python.keras.models import Sequential
    from tensorflow.python.keras.layers import Conv2D, MaxPooling2D, BatchNormalization
    from tensorflow.python.keras.layers import Activation, Dropout, Flatten, Dense

\#Каталог с данными для обучения  
train\_dir = ‘train’

# Каталог с данными для проверки #

val\_dir = ‘val’

# Каталог с данными для тестирования #

test\_dir = ‘val’

# Размеры изображения #

img\_width, img\_height = 800, 800

# Размерность тензора на основе изображения для входных данных в нейронную сеть #

# backend Tensorflow, channels\_last #

input\_shape = (img\_width, img\_height, 3)

# Количество эпох #

epochs = 1

# Размер мини-выборки #

batch\_size = 4

# Количество изображений для обучения #

nb\_train\_samples = 300

# Количество изображений для проверки #

nb\_validation\_samples = 25

# Количество изображений для тестирования #

nb\_test\_samples = 25

model = Sequential()

model.add(Conv2D(32, (7, 7), padding=“same”, input\_shape=input\_shape))  
model.add(BatchNormalization())  
model.add(Activation(‘tanh’))  
model.add(MaxPooling2D(pool\_size=(10, 10)))

model.add(Conv2D(64, (5, 5), padding=“same”))  
model.add(BatchNormalization())  
model.add(Activation(‘tanh’))  
model.add(MaxPooling2D(pool\_size=(10, 10)))

model.add(Flatten())  
model.add(Dense(512))  
model.add(Activation(‘relu’))  
model.add(Dropout(0.5))  
model.add(Dense(10, activation=‘softmax’))

model.compile(loss=‘categorical\_crossentropy’,  
optimizer=“Nadam”,  
metrics=\[‘accuracy’\])  
print(model.summary())  
datagen = ImageDataGenerator(rescale=1. / 255)

train\_generator = datagen.flow\_from\_directory(  
train\_dir,  
target\_size=(img\_width, img\_height),  
batch\_size=batch\_size,  
class\_mode=‘categorical’)

val\_generator = datagen.flow\_from\_directory(  
val\_dir,  
target\_size=(img\_width, img\_height),  
batch\_size=batch\_size,  
class\_mode=‘categorical’)

test\_generator = datagen.flow\_from\_directory(  
test\_dir,  
target\_size=(img\_width, img\_height),  
batch\_size=batch\_size,  
class\_mode=‘categorical’)

model.fit\_generator(  
train\_generator,  
steps\_per\_epoch=nb\_train\_samples // batch\_size,  
epochs=epochs,  
validation\_data=val\_generator,  
validation\_steps=nb\_validation\_samples // batch\_size)

print(‘Сохраняем сеть’)  
model.save(“grib.h5”)  
print(“Сохранение завершено!”)

模型载入

print(“Загрузка сети”)  
model = load\_model(“grib.h5”)  
print(“Загрузка завершена!”)

报错

/usr/bin/python3.5 /home/disk2/py/neroset/do.py
    /home/mama/.local/lib/python3.5/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
     from ._conv import register_converters as _register_converters
    Using TensorFlow backend.
    Загрузка сети
    Traceback (most recent call last):
     File "/home/disk2/py/neroset/do.py", line 13, in <module>
      model = load_model("grib.h5")
     File "/usr/local/lib/python3.5/dist-packages/keras/models.py", line 243, in load_model
      model = model_from_config(model_config, custom_objects=custom_objects)
     File "/usr/local/lib/python3.5/dist-packages/keras/models.py", line 317, in model_from_config
      return layer_module.deserialize(config, custom_objects=custom_objects)
     File "/usr/local/lib/python3.5/dist-packages/keras/layers/__init__.py", line 55, in deserialize
      printable_module_name='layer')
     File "/usr/local/lib/python3.5/dist-packages/keras/utils/generic_utils.py", line 144, in deserialize_keras_object
      list(custom_objects.items())))
     File "/usr/local/lib/python3.5/dist-packages/keras/models.py", line 1350, in from_config
      model.add(layer)
     File "/usr/local/lib/python3.5/dist-packages/keras/models.py", line 492, in add
      output_tensor = layer(self.outputs[0])
     File "/usr/local/lib/python3.5/dist-packages/keras/engine/topology.py", line 590, in __call__
      self.build(input_shapes[0])
     File "/usr/local/lib/python3.5/dist-packages/keras/layers/normalization.py", line 92, in build
      dim = input_shape[self.axis]
    TypeError: tuple indices must be integers or slices, not list

Process finished with exit code 1

战斗种族解释

убераю BatchNormalization всё работает хорошо. Не подскажите в чём ошибка?Выяснил что сохранение keras и нормализация tensorflow не работают вместе нужно просто изменить строку импорта.（译文：整理BatchNormalization一切正常。 不要告诉我错误是什么？我发现保存keras和规范化tensorflow不能一起工作;只需更改导入字符串即可。)

**强调文本 强调文本**

keras.preprocessing.image import ImageDataGenerator
    keras.models import Sequential
    keras.layers import Conv2D, MaxPooling2D, BatchNormalization
    keras.layers import Activation, Dropout, Flatten, Dense

\#\#完美解决

\#\#附上原文链接

> https://qa-help.ru/questions/keras-batchnormalization

以上这篇keras的ImageDataGenerator和flow()的用法说明就是小编分享给大家的全部内容了，希望能给大家一个参考，也希望大家多多支持菜鸟教程www.piaodoo.com。

[Link 1]: https://www.piaodoo.com/
[Link 2]: https://www.hnthzk.com/
[Link 3]: http://www.op-kg.com/