Ding Chen
In GAIL, I tested this hyper parameters and this seems to work a lot more consistently: a2c_hidden_size = 32 discrim_hidden_size = 128 lr = 1e-3 num_steps = 128 mini_batch_size =...
Ubuntu 18.04.5 LTS Meet this problem after adding 'mp.set_start_method('spawn')' https://github.com/tristandeleu/pytorch-maml-rl/issues/40#issuecomment-632598191 Traceback (most recent call last): File "train.py", line 122, in main(args) File "train.py", line 54, in main num_workers=args.num_workers) File "/home/dchen/pytorch-maml-rl/maml_rl/samplers/multi_task_sampler.py",...
tensorflow.python.framework.errors_impl.NotFoundError: /home/csu/anaconda3/envs/tf/lib/python3.6/site-packages/ray/pyarrow_files/pyarrow/tensorflow/plasma_op.so: cannot open shared object file: No such file or directory
According to the source code, the only engineer who implemented multiple actions overridden the act method and seemed to not need react
The paper mentions ‘We implement Double DQN [72] and Prioritized Experience Replay [58] on four Atari games’, but the actual code implementation does not implement Double DQN, but rather Dueling...
(eprop-figure) rlcube@rlcube:~/dchen/eligibility_propagation/Figure_4_and_5_ATARI$ python compile.py ale - compiling using command: g++ -std=c++11 -shared ale.cc -o tfaleop.so -fPIC -I /home/rlcube/miniconda3/envs/eprop-figure/lib/python3.6/site-packages/tensorflow_core/include -O2 -D_GLIBCXX_USE_CXX11_ABI=1 -L/home/rlcube/miniconda3/envs/eprop-figure/lib/python3.6/site-packages/tensorflow_core -I/home/rlcube/miniconda3/envs/eprop-figure/lib/python3.6/site-packages/tensorflow_core/include -D_GLIBCXX_USE_CXX11_ABI=1 -L/home/rlcube/miniconda3/envs/eprop-figure/lib/python3.6/site-packages/tensorflow_core -l:libtensorflow_framework.so.1 -Iale/include -Lale/lib -lale In file...