From aa6e58bdf13b8cb1fdf385060a98454fa05cf1b0 Mon Sep 17 00:00:00 2001 From: John Schulman Date: Sun, 27 Aug 2017 22:22:14 -0700 Subject: [PATCH] fix readmes --- README.md | 2 +- baselines/ppo1/README.md | 2 +- baselines/trpo_mpi/README.md | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index bd929b2..e041600 100644 --- a/README.md +++ b/README.md @@ -18,5 +18,5 @@ pip install -e . - [ACKTR](baselines/acktr) - [DDPG](baselines/ddpg) - [DQN](baselines/deepq) -- [PPO](baselines/pposgd) +- [PPO](baselines/ppo1) - [TRPO](baselines/trpo_mpi) diff --git a/baselines/ppo1/README.md b/baselines/ppo1/README.md index 6ef3cbe..70325b7 100644 --- a/baselines/ppo1/README.md +++ b/baselines/ppo1/README.md @@ -2,6 +2,6 @@ - Original paper: https://arxiv.org/abs/1707.06347 - Baselines blog post: https://blog.openai.com/openai-baselines-ppo/ -- `python -m baselines.ppo.run_atari` runs the algorithm for 40M frames = 10M timesteps on an Atari game. See help (`-h`) for more options. +- `mpirun -np 8 python -m baselines.ppo.run_atari` runs the algorithm for 40M frames = 10M timesteps on an Atari game. See help (`-h`) for more options. - `python -m baselines.ppo.run_mujoco` runs the algorithm for 1M frames on a Mujoco environment. diff --git a/baselines/trpo_mpi/README.md b/baselines/trpo_mpi/README.md index 07ae7dd..c7e379d 100644 --- a/baselines/trpo_mpi/README.md +++ b/baselines/trpo_mpi/README.md @@ -2,5 +2,5 @@ - Original paper: https://arxiv.org/abs/1502.05477 - Baselines blog post https://blog.openai.com/openai-baselines-ppo/ -- `python -m baselines.ppo1.run_atari` runs the algorithm for 40M frames = 10M timesteps on an Atari game. See help (`-h`) for more options. +- `mpirun -np 16 python -m baselines.ppo1.run_atari` runs the algorithm for 40M frames = 10M timesteps on an Atari game. See help (`-h`) for more options. - `python -m baselines.ppo1.run_mujoco` runs the algorithm for 1M timesteps on a Mujoco environment. \ No newline at end of file