2

我是 tensorflow 的新手，我正在尝试建立一个具有密集翻转层的贝叶斯神经网络。我的代码如下所示：

from tensorflow.keras.models import Sequential
import tensorflow_probability as tfp
import tensorflow as tf


def train_BNN(training_data, training_labels, test_data, test_labels, layers, epochs):

 bayesian_nn = Sequential()
 nbr_samples, dim = training_data.shape
 training_labels = training_labels.reshape((nbr_samples, 1))
 training_data = tf.convert_to_tensor(training_data, tf.float32)
 training_labels = tf.convert_to_tensor(training_labels, tf.float32)
 for i in range(0, len(layers)):
    bayesian_nn.add(tfp.layers.DenseFlipout(layers[i], activation='relu'))
 bayesian_nn.add(tfp.layers.DenseFlipout(1, activation='sigmoid'))

# Define loss function
 logits = bayesian_nn(training_data)
 neg_log_likelihood = tf.losses.sigmoid_cross_entropy(training_labels, logits=logits)
 kl = sum(bayesian_nn.losses)
 loss = neg_log_likelihood + kl
 # Define optimizer
 train_op = tf.train.AdamOptimizer().minimize(loss)
 bayesian_nn.compile(train_op, loss=loss, metrics=['accuracy'])

尝试编译网络时出现以下错误：

Traceback (most recent call last):
  File "./test_script.py", line 34, in <module>
    dataset[0:training,], labels[0:training], dataset[training:,], labels[training:], layers, epochs
  File "/home/e/bayesian_nn.py", line 34, in train_BNN
    bayesian_nn.compile(train_op, loss=loss, metrics=['accuracy'])
  File "/home/e/.pyenv/versions/kmer_profiling/lib/python3.6/site-packages/tensorflow/python/training/checkpointable/base.py", line 474, in _method_wrapper
    method(self, *args, **kwargs)
  File "/home/e/.pyenv/versions/kmer_profiling/lib/python3.6/site-packages/tensorflow/python/keras/engine/training.py", line 405, in compile
    loss = loss or {}
  File "/home/e/.pyenv/versions/kmer_profiling/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 671, in __bool__
    raise TypeError("Using a `tf.Tensor` as a Python `bool` is not allowed. "
TypeError: Using a `tf.Tensor` as a Python `bool` is not allowed. Use `if t is not None:` instead of `if t:` to test if a tensor is defined, and use TensorFlow ops such as tf.cond to execute subgraphs conditioned on the value of a tensor.

我不知道是什么导致了这个错误......请帮助:)

4

1 回答 1

0

loss您传入的似乎model.compile是 a tf.Tensor，这是不允许的。查看文档，您似乎应该直接传递损失函数（或字符串表示），而不是先实例化它，即loss=tf.losses.sigmoid_cross_entropy（或损失列表）而不是loss=tf.losses.sigmoid_cross_entropy(training_labels, logits=logits)

于 2019-03-08T23:38:03.803 回答