Open ghost opened 4 years ago
code-terminator
commented
4 years ago Hi Julia,
It seems you the error comes from you do not have a GPU.
In terms of the error you mentioned in the email, please check your version of Tensorflow and system requirement in README.
srhthu
commented
2 years ago I have met the same issue. The reason is the corresponding version of cuda and cudnn is not installed. To solve it, just run: conda install cudatoolkit=10.0 conda install cudnn
Hi I am trying to run the codes with the specified versions. got this error. thanks for your help
WARNING:tensorflow: The TensorFlow contrib module will not be included in TensorFlow 2.0. For more information, please see:
Traceback (most recent call last): File "imdb_demo.py", line 162, in
global_step = train_imdb(D_tr, inv_rnn, opts, global_step, args)
File "/remote/svm/user.active/julia/dev/invariant_rationalization/train.py", line 30, in train_imdb
inputs, masks, envs)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/keras/engine/base_layer.py", line 679, in call
outputs = self.call(inputs, *args, kwargs)
File "/remote/svm/user.active/julia/dev/invariant_rationalization/model.py", line 130, in call
gen_outputs = self.generator(gen_embeddings)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/keras/engine/base_layer.py", line 679, in call
outputs = self.call(inputs, args, kwargs)
File "/remote/.svm/user.active/julia/dev/invariant_rationalization/model.py", line 59, in call
h = self.rnn(x)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/keras/layers/wrappers.py", line 533, in call
return super(Bidirectional, self).call(inputs, kwargs)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/keras/engine/base_layer.py", line 679, in call
outputs = self.call(inputs, args, kwargs)
File "//user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/keras/layers/wrappers.py", line 633, in call
y = self.forward_layer.call(inputs, **kwargs)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/keras/layers/cudnn_recurrent.py", line 110, in call
output, states = self._process_batch(inputs, initial_state)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/keras/layers/cudnn_recurrent.py", line 302, in _process_batch
rnn_mode='gru')
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/ops/gen_cudnn_rnn_ops.py", line 109, in cudnn_rnn
ctx=_ctx)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/ops/gen_cudnn_rnn_ops.py", line 197, in cudnn_rnn_eager_fallback
attrs=_attrs, ctx=_ctx, name=name)
File "/user/julia/libs/anaconda3/envs/irm/lib/python3.6/site-packages/tensorflow/python/eager/execute.py", line 67, in quick_execute
six.raise_from(core._status_to_exception(e.code, message), None)
File "", line 3, in raise_from
tensorflow.python.framework.errors_impl.InternalError: Could not find valid device for node.
Node: {{node CudnnRNN}}
All kernels registered for op CudnnRNN :
device='GPU'; T in [DT_DOUBLE]
device='GPU'; T in [DT_FLOAT]
device='GPU'; T in [DT_HALF]
[Op:CudnnRNN]