python - 无属性 'get_default_graph'

Question

我有一个 python 代码:

from keras.models import Sequential
....
gc.collect()
model = Sequential()
model.add(Dense(32,input_shape = (6,),activation = 'relu'))
...

但是它在行 model=Sequential() 中给我一个错误。它是:

AttributeError: 'module' 对象没有属性 'get_default_graph'

或完整形式:

File "neuralnetwork.py", line 38, in <module> model = Sequential()
File "/usr/local/lib/python2.7/dist-packages/keras/engine/sequential.py", line 87, in __init__
super(Sequential, self).__init__(name=name)
File "/usr/local/lib/python2.7/dist-packages/keras/legacy/interfaces.py", line 91, in wrapper
return func(*args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/keras/engine/network.py", line 96, in __init__
self._init_subclassed_network(**kwargs)
File "/usr/local/lib/python2.7/dist-packages/keras/engine/network.py", line 294, in _init_subclassed_network
self._base_init(name=name)
File "/usr/local/lib/python2.7/dist-packages/keras/engine/network.py", line 109, in _base_init
name = prefix + '_' + str(K.get_uid(prefix))
File "/usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py", line 74, in get_uid
graph = tf.get_default_graph()
AttributeError: 'module' object has no attribute 'get_default_graph'

我确实了解神经网络，了解它们的工作原理等等。我已经从头开始实现了一个。但我是第一次使用 tensorflow，不知道这个错误可能意味着什么。请帮我解决这个问题。谢谢!

Answer 1

由于您使用的是 tensorflow 2.0，因此您应该在 tensorflow 中使用 keras API。将导入语句更改为:

import tensorflow as tf
from tensorflow import keras
....
model = keras.Sequential()
model.add(keras.layers.Dense(32,input_shape = (6,),activation = 'relu'))
...