effective_transformer
effective_transformer copied to clipboard
No OpKernel was registered to support Op 'EffectiveTransformerInput' used by node EffectiveTransformerInput
I build the effective_transformer successfully, but when I run the benchmark.py, get the error, how can I fix the error
Traceback (most recent call last): File "/home/luban/miniconda3/envs/study/lib/python3.7/site-packages/tensorflow_core/python/client/session.py", line 1365, in _do_call return fn(*args) File "/home/luban/miniconda3/envs/study/lib/python3.7/site-packages/tensorflow_core/python/client/session.py", line 1348, in _run_fn self._extend_graph() File "/home/luban/miniconda3/envs/study/lib/python3.7/site-packages/tensorflow_core/python/client/session.py", line 1388, in _extend_graph tf_session.ExtendSession(self._session) tensorflow.python.framework.errors_impl.InvalidArgumentError: No OpKernel was registered to support Op 'EffectiveTransformerInput' used by {{node EffectiveTransformerInput}}with these attrs: [batch_size=1, to_seq_len=128, T=DT_FLOAT, head_num=12, from_seq_len=128, size_per_head=64] Registered devices: [CPU, XLA_CPU, XLA_GPU] Registered kernels: device='GPU'; T in [DT_HALF] device='GPU'; T in [DT_FLOAT]
[[EffectiveTransformerInput]]
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "benchmark.py", line 142, in
[[EffectiveTransformerInput]]
please check your tf gpu environment
hi @wwy510553871, I got similar error here, did you manage to resolve this issue?