Question

我有以下代码：

_X = np.arange(1, 7).reshape((2, 3))
_Y = np.arange(1, 7).reshape((3, 2))

X = tf.convert_to_tensor(_X)
Y = tf.convert_to_tensor(_Y)

# Matrix multiplication
out1 = tf.matmul(X, Y)

对于它，我收到了这个错误：

TypeError: Value passed to parameter 'a' has DataType int64 not in list of allowed values: float16, float32, float64, int32, complex64, complex128

我使用的是最新版本的Tensorflow。可能是什么问题？

Answer 1

输入到tf.matmul只接受这些dtypes：

a: Tensor of type float16, float32, float64, int32, complex64, complex128 and rank > 1.

将X和Y的dtype更改为上面的dtypes。

import tensorflow as tf
import numpy as np
_X = np.arange(1, 7).reshape((2, 3))
_Y = np.arange(1, 7).reshape((3, 2))

X = tf.convert_to_tensor(_X,dtype=tf.int32)
Y = tf.convert_to_tensor(_Y,dtype=tf.int32)

# Matrix multiplication
out1 = tf.matmul(X, Y)

sess = tf.Session()
print(sess.run(out1))

Answer 2

您可以尝试这种方式：

def weighted_binary_crossentropy(y_true, y_pred):
   y_true = tensorflow.cast(y_true, tensorflow.float32)
   y_pred = tensorflow.cast(y_pred, tensorflow.float32)
   ...

TypeError：传递给参数'a'的值使DataType不在允许值列表中

2 个答案: