You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I finally got TF working and tried to run the alpha zero examples. Running both alpha_zero.py on Quoridor and running the tic_tac_toe_alpha_zero.py yielded the same failure. Is this something known? Is it possibly a library mismatch (I have to say that the version management for Cuda/CudNN/TF/etc is pretty confusing for a newbie, so I'm sure I could have messed something up.)
I will continue to poke around and see if I can figure it out.
File "/home/certain/src/open_spiel/test/../open_spiel/python/examples/tic_tac_toe_alpha_zero.py", line 68, in <module>
app.run(main)
File "/home/certain/src/open_spiel/venv/lib/python3.10/site-packages/absl/app.py", line 308, in run
_run_main(main, args)
File "/home/certain/src/open_spiel/venv/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main
sys.exit(main(argv))
File "/home/certain/src/open_spiel/test/../open_spiel/python/examples/tic_tac_toe_alpha_zero.py", line 63, in main
alpha_zero.alpha_zero(config)
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/alpha_zero.py", line 545, in alpha_zero
learner(game=game, config=config, actors=actors, # pylint: disable=missing-kwoa
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/alpha_zero.py", line 171, in _watcher
return fn(config=config, logger=logger, **kwargs)
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/alpha_zero.py", line 331, in learner
model = _init_model_from_config(config)
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/alpha_zero.py", line 148, in _init_model_from_config
return model_lib.Model.build_model(
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/model.py", line 173, in build_model
cls._define_graph(model_type, input_shape, output_size, nn_width,
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/model.py", line 241, in _define_graph
torso = cascade(observations, [
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/model.py", line 28, in cascade
x = fn(x)
File "/home/certain/src/open_spiel/open_spiel/python/algorithms/alpha_zero/model.py", line 50, in batch_norm_layer
applied = bn(x, training)
File "/home/certain/src/open_spiel/venv/lib/python3.10/site-packages/keras/src/utils/traceback_utils.py", line 122, in error_handler
raise e.with_traceback(filtered_tb) from None
File "/home/certain/src/open_spiel/venv/lib/python3.10/site-packages/tensorflow/python/framework/tensor_shape.py", line 1440, in as_list
raise ValueError("as_list() is not defined on an unknown TensorShape.")
ValueError: as_list() is not defined on an unknown TensorShape.
The text was updated successfully, but these errors were encountered:
I finally got TF working and tried to run the alpha zero examples. Running both alpha_zero.py on Quoridor and running the tic_tac_toe_alpha_zero.py yielded the same failure. Is this something known? Is it possibly a library mismatch (I have to say that the version management for Cuda/CudNN/TF/etc is pretty confusing for a newbie, so I'm sure I could have messed something up.)
I will continue to poke around and see if I can figure it out.
The text was updated successfully, but these errors were encountered: