search

google-research / dads

Code for 'Dynamics-Aware Unsupervised Discovery of Skills' (DADS). Enables skill discovery without supervision, which can be combined with model-based control.
Apache License 2.0
186 stars 51 forks source link

Check failed: work_element_count > 0 #2

Closed shahdev closed 4 years ago

shahdev commented 4 years ago

Hello Archit, thanks for the code!!

I downloaded the dads repository and created a conda environment using the env.yml file. When I run this command: python unsupervised_skill_learning/dads_off.py --logdir=logs/ --flagfile=configs/ant_xy_onpolicy.txt (after changing the Ant environment to Ant-v3)

I get this error: 2020-03-24 14:20:44.165003: F ./tensorflow/core/util/gpu_launch_config.h:129] Check failed: work_element_count > 0 (0 vs. 0) Fatal Python error: Aborted

Thread 0x00007f67acff9700 (most recent call first): File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/threading.py", line 295 in wait File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/queue.py", line 164 in get File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/summary/writer/event_file_writer.py", line 159 in run File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/threading.py", line 916 in _bootstrap_inner File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/threading.py", line 884 in _bootstrap

Thread 0x00007f71d3555740 (most recent call first): File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1445 in _call_tf_sessionrun File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1352 in _run_fn File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1367 in _do_call File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1361 in _do_run File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1183 in _run File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 960 in run File "/data/dpshah2/SaurabhG/dads/unsupervised_skill_learning/skill_dynamics.py", line 399 in train File "unsupervised_skill_learning/dads_off.py", line 1406 in main File "/home/dpshah2/.local/lib/python3.6/site-packages/absl/app.py", line 250 in _run_main File "/home/dpshah2/.local/lib/python3.6/site-packages/absl/app.py", line 299 in run File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/platform/app.py", line 40 in run File "unsupervised_skill_learning/dads_off.py", line 1711 in Aborted (core dumped)

Could you help out here!! Thank you :)

architsharma97 commented 4 years ago

Could you consider running on CPU only? (it will probably be faster too as the neural networks are really small)

On Tue, Mar 24, 2020 at 12:24 PM Devansh Shah notifications@github.com wrote:

Hello Archit, thanks for the code!!

I downloaded the dads repository and created a conda environment using the env.yml file. When I run this command: python unsupervised_skill_learning/dads_off.py --logdir=logs/ --flagfile=configs/ant_xy_onpolicy.txt (after changing the Ant environment to Ant-v3)

I get this error: 2020-03-24 14:20:44.165003: F ./tensorflow/core/util/gpu_launch_config.h:129] Check failed: work_element_count > 0 (0 vs. 0) Fatal Python error: Aborted

Thread 0x00007f67acff9700 (most recent call first): File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/threading.py", line 295 in wait File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/queue.py", line 164 in get File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/summary/writer/event_file_writer.py", line 159 in run File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/threading.py", line 916 in _bootstrap_inner File "/home/dpshah2/miniconda3/envs/dads-env/lib/python3.6/threading.py", line 884 in _bootstrap

Thread 0x00007f71d3555740 (most recent call first): File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1445 in _call_tf_sessionrun File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1352 in _run_fn File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1367 in _do_call File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1361 in _do_run File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1183 in _run File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 960 in run File "/data/dpshah2/SaurabhG/dads/unsupervised_skill_learning/skill_dynamics.py", line 399 in train File "unsupervised_skill_learning/dads_off.py", line 1406 in main File "/home/dpshah2/.local/lib/python3.6/site-packages/absl/app.py", line 250 in _run_main File "/home/dpshah2/.local/lib/python3.6/site-packages/absl/app.py", line 299 in run File "/home/dpshah2/.local/lib/python3.6/site-packages/tensorflow_core/python/platform/app.py", line 40 in run File "unsupervised_skill_learning/dads_off.py", line 1711 in Aborted (core dumped)

Could you help out here!! Thank you :)

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/google-research/dads/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADGCUKAIV5LHC7O4PJEL67LRJECFRANCNFSM4LS5WF3Q .

shahdev commented 4 years ago

Oh great yes it is running now on CPU. Thanks a lot!! Yes cpu will be faster too because the main bottle-neck would be stepping through the environment which does not particularly require a gpu.