InvalidArgumentError of run_CIL.py
gavin8533 opened this issue · 1 comments
Hi @felipecode
I suffer from an issue when running run_CIL.py of invalid argument error as:
InvalidArgumentError (see above for traceback): Cannot assign a device for operation input_image: node input_image (defined at /home/XXXXX/imitation-learning/agents/imitation/imitation_learning.py:42) was explicitly assigned to /device:GPU:0 but available devices are [ /job:localhost/replica:0/task:0/device:CPU:0, /job:localhost/replica:0/task:0/device:XLA_CPU:0 ]. Make sure the device specification refers to a valid device. The requested device appears to be a GPU, but CUDA is not enabled.
[[node input_image (defined at /home/XXXXX/imitation-learning/agents/imitation/imitation_learning.py:42) ]]
Thanks in advance.
Problem fixed by upgrade CUDA to 9.0 and uninstall and install tensorflow-gpu.
Issue close!