Open foxnudt opened 7 years ago
I get the same error message. Could you solve it? How?
Just change the backend to Theano. This implementation was built using Theano, so it's not guarantee to work with TF.
To change it, you can follow the instructions here.
yes, thx
File "/Users/fox/PycharmProjects/dqn-master/dqn.py", line 53, in build_functions self.build_model()
assert type(outputs) in {list, tuple}, 'Output to a TensorFlow backend function should be a list or tuple.' AssertionError: Output to a TensorFlow backend function should be a list or tuple.