属性错误：模块'keras.engine.base_layer'没有属性'BaseRandomLayer'

Question

上下文-我在保存/加载EfficientNetB7时出现错误，我看到的一篇文章建议每晚安装tf。这真的把我的Google Colab搞砸了!我卸载了tf-nightly，重新启动了运行时，现在当我去运行导入单元时，我得到了下面的错误:

我重新安装了tensorflow 2.6.0

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-25-5d853ee59f92> in <module>()
11 import tensorflow as tf
12 from tensorflow import keras
---> 13 from tensorflow.keras import Model
14 from tensorflow.keras.models import Sequential
15 from tensorflow.keras.utils import to_categorical
11 frames
/usr/local/lib/python3.7/dist-packages/keras/layers/core/dropout.py in <module>()
24 
25 @keras_export('keras.layers.Dropout')
---> 26 class Dropout(base_layer.BaseRandomLayer):
27   """Applies Dropout to the input.
28 
AttributeError: module 'keras.engine.base_layer' has no attribute 'BaseRandomLayer'

我之前也遇到过这个导入的问题:

from tensorflow.keras import backend as keras

这是我所有的进口:

import os
import glob
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from skimage.io import imread
from skimage.transform import resize
from sklearn.preprocessing import OneHotEncoder
from sklearn.model_selection import train_test_split
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import Model
from tensorflow.keras.models import Sequential
from tensorflow.keras.utils import to_categorical
from tensorflow.keras.losses import categorical_crossentropy
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D, Dropout, BatchNormalization
from tensorflow.keras.callbacks import ModelCheckpoint, LearningRateScheduler
from datetime import datetime
from sklearn.metrics import confusion_matrix, accuracy_score, f1_score, precision_score, recall_score
import skimage.io as io
import skimage.transform as trans
from tensorflow.keras import models
from tensorflow.keras import layers
from tensorflow.keras import optimizers
from tensorflow.keras.models import *
from tensorflow.keras.layers import *
from tensorflow.keras.optimizers import *
from tensorflow.keras.callbacks import ModelCheckpoint, LearningRateScheduler
from tensorflow.keras import backend as keras
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from keras_preprocessing.image import ImageDataGenerator
from tensorflow.keras.optimizers.schedules import ExponentialDecay
import scipy.misc as sc
import h5py
from sklearn.semi_supervised import LabelSpreading
from sklearn.model_selection import GridSearchCV
from tensorflow.keras.applications import * #Efficient Net included here
import shutil
from sklearn import model_selection
from tqdm import tqdm
# import warnings
# warnings.filterwarnings("ignore", category=DeprecationWarning)

Answer 1

尝试回滚python版本到3.6。如果使用Anaconda，请运行

conda install python=3.6

然后安装keras

pip install keras==2.0.8

在Ubuntu中为我工作

Answer 2

安装tensorflow==2.8.0后，我能够摆脱这个问题。