如何修复TensorFlow中的ValueError ?

Question

我用TensorFlow尝试了一些东西，得到了这个错误:

ValueError: Input 0 of layer gru is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: (None, 3)

如何解决这个问题?

Answer 1

我可以用示例代码重现您的问题。tf.keras.layers.GRU期望输入A三维张量，形状为[batch, timesteps, feature]

<<p>复制代码/strong>

import tensorflow as tf
inputs = tf.random.normal([32, 8])
gru = tf.keras.layers.GRU(4)
output = gru(inputs)
print(output.shape)

ValueError: Input 0 of layer gru_1 is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: (32, 8)

工作样例代码

import tensorflow as tf
inputs = tf.random.normal([32, 10, 8])
gru = tf.keras.layers.GRU(4)
output = gru(inputs)
print(output.shape)

(32, 4)