janus-shiau / awd-lstm-tensorflow Goto Github PK

When I modified x to 3 dimension:[batch_size, time_step, hidden_size]， the program seems to not work:
Traceback (most recent call last):
File "example_awd_lstm.py", line 158, in
test_with_sess_run(x)
File "example_awd_lstm.py", line 60, in test_with_sess_run
x_out, _ = lstm_cell(x, init_states)
File "/home/punjiujam/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/ops/rnn_cell_impl.py", line 371, in call
*args, **kwargs)
File "/home/punjiujam/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/layers/base.py", line 530, in call
outputs = super(Layer, self).call(inputs, *args, **kwargs)
File "/home/punjiujam/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/keras/engine/base_layer.py", line 538, in call
self._maybe_build(inputs)
File "/home/punjiujam/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/keras/engine/base_layer.py", line 1591, in _maybe_build
self.input_spec, inputs, self.name)
File "/home/punjiujam/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/keras/engine/input_spec.py", line 123, in assert_input_compatibility
str(x.shape.as_list()))
ValueError: Input 0 of layer weight_drop_lstm_cell_1 is incompatible with the layer: expected ndim=2, found ndim=3. Full shape received: [2, 10, 10]