tensorflow - keras 模型适合 : ValueError: Failed to find data adapter that can handle input: , <类 'NoneType' >

Question

我正在为多类分类构建一个简单的 CNN 模型。根据ImageDataGenerator的flow_from_directory函数的要求，训练和测试数据在data_path中。

这是我根据数据构建和训练模型的代码:

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

# Build Model

model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(40, 24, 1)))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
model.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
model.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(12, activation='softmax'))

model.compile('binary_crossentropy', 'SGD', ['accuracy'])

# Init Generators

generator = ImageDataGenerator(rescale=1./255,
                               horizontal_flip=True,
                               fill_mode='nearest',
                               validation_split=0.2)

def get_train_images():
    train_images = generator.flow_from_directory(os.path.join(data_path, 'train'),
                                                 target_size=(40, 24, 1),
                                                 batch_size=32,
                                                 color_mode='grayscale',
                                                 class_mode='categorical',
                                                 subset='training',
                                                 shuffle=True)

def get_validation_images():
    validation_images = generator.flow_from_directory(os.path.join(data_path, 'train'),
                                                      target_size=(40, 24, 1),
                                                      batch_size=32,
                                                      color_mode='grayscale',
                                                    class_mode='categorical',
                                                      subset='validation',
                                                      shuffle=True)

# Train Model

model.fit(get_train_images, validation_data=get_validation_images, epochs=20)

拟合函数给出以下错误:

File "C:\Python38\lib\site-packages\tensorflow\python\keras\engine\training.py", line 108, in _method_wrapper
    return method(self, *args, **kwargs)
  File "C:\Python38\lib\site-packages\tensorflow\python\keras\engine\training.py", line 1049, in fit
    data_handler = data_adapter.DataHandler(
  File "C:\Python38\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py", line 1104, in __init__
    adapter_cls = select_data_adapter(x, y)
  File "C:\Python38\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py", line 968, in select_data_adapter
    raise ValueError(
ValueError: Failed to find data adapter that can handle input: <class 'method'>, <class 'NoneType'>

看起来是某种兼容性问题。我正在使用 tensorflow 版本 2.3.1。有人可以指出我做错了什么并帮助我解决这个问题吗？

谢谢!

Answer 1

为了解决这个问题，我必须更改两件事:

• flow_from_directory 的目标大小应该是 (40, 24) 而不是 (40, 24, 1)
• 我有函数包装器来获取 flow_from_directory 生成器，我将这些函数作为参数传递给 fit 函数。相反，我必须将这些包装器的返回值传递给 fit 函数

正确的做法应该是:

model.fit(get_train_images(), validation_data=get_validation_images(), epochs=20)