Python TensorFlow：将 float64 张量转换为 float32
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/35725513/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
TensorFlow: cast a float64 tensor to float32
提问by Karishma Malkan
I am trying to use: train = optimizer.minimize(loss)but the standard optimizers do not work with tf.float64. Therefore I want to truncate my lossfrom tf.float64to only tf.float32.
我正在尝试使用：train = optimizer.minimize(loss)但标准优化器不适用于tf.float64. 因此，我想我的截断loss从tf.float64只tf.float32。
Traceback (most recent call last):
File "q4.py", line 85, in <module>
train = optimizer.minimize(loss)
File "/Library/Python/2.7/site-packages/tensorflow/python/training/optimizer.py", line 190, in minimize
colocate_gradients_with_ops=colocate_gradients_with_ops)
File "/Library/Python/2.7/site-packages/tensorflow/python/training/optimizer.py", line 229, in compute_gradients
self._assert_valid_dtypes([loss])
File "/Library/Python/2.7/site-packages/tensorflow/python/training/optimizer.py", line 354, in _assert_valid_dtypes
dtype, t.name, [v for v in valid_dtypes]))
ValueError: Invalid type tf.float64 for Add_1:0, expected: [tf.float32].
回答by mrry
The short answer is that you can convert a tensor from tf.float64to tf.float32using the tf.cast()op:
简短的回答是，你可以将张量从转换tf.float64到tf.float32使用tf.cast()OP：
loss = tf.cast(loss, tf.float32)
The longer answer is that this will not solve all of your problems with the optimizers. (The lack of support for tf.float64is a known issue.) The optimizers require that all of the tf.Variableobjects that you are trying to optimize must also have type tf.float32.
更长的答案是，这不会解决优化器的所有问题。（缺乏对 的支持tf.float64是一个已知问题。）优化器要求tf.Variable您尝试优化的所有对象也必须具有 type tf.float32。
