Hi, when running lstm.py, just after loading the midis I get this error:
Traceback (most recent call last):
File "lstm.py", line 176, in <module>
train_network()
File "lstm.py", line 26, in train_network
model = create_network(network_input, n_vocab)
File "lstm.py", line 103, in create_network
model.add(CuDNNLSTM(512,input_shape=(network_input.shape[1], network_input.shape[2]),return_sequences=True))
File "/Users/ec/pypi_venv/lib/python3.7/site-packages/keras/engine/sequential.py", line 166, in add
layer(x)
File "/Users/ec/pypi_venv/lib/python3.7/site-packages/keras/layers/recurrent.py", line 541, in __call__
return super(RNN, self).__call__(inputs, **kwargs)
File "/Users/ec/pypi_venv/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py", line 75, in symbolic_fn_wrapper
return func(*args, **kwargs)
File "/Users/ec/pypi_venv/lib/python3.7/site-packages/keras/engine/base_layer.py", line 463, in __call__
self.build(unpack_singleton(input_shapes))
File "/Users/ec/pypi_venv/lib/python3.7/site-packages/keras/layers/cudnn_recurrent.py", line 425, in build
from tensorflow.contrib.cudnn_rnn.python.ops import cudnn_rnn_ops
ModuleNotFoundError: No module named 'tensorflow.contrib'
I have the following packages installed via pip. Are they not the correct version?
tensorflow 2.0.0
tensorflow-estimator 2.0.1
I think having a requirements.txt might help if that's the case. Thanks!
Hi, when running
lstm.py
, just after loading the midis I get this error:I have the following packages installed via pip. Are they not the correct version?
I think having a requirements.txt might help if that's the case. Thanks!