├── .coveragerc ├── .github └── workflows │ └── coverage.yml ├── .gitignore ├── .gitmodules ├── LICENSE ├── README.md ├── example ├── README.md ├── atari │ ├── README.md │ ├── train_dqn.py │ └── train_ppo.py ├── benchmark.md ├── cartpole │ ├── train_dqn.py │ └── train_ppo.py └── pybullet │ └── README.md ├── experiments ├── __init__.py ├── base.py └── ppo │ ├── __init__.py │ ├── config.py │ └── run.py ├── setup.cfg ├── setup.py ├── test ├── __init__.py ├── test_algo │ ├── __init__.py │ ├── test_dqn.py │ └── test_ppo.py ├── test_lib │ ├── __init__.py │ ├── test_base.py │ ├── test_data.py │ ├── test_envs │ │ ├── __init__.py │ │ ├── test_atari.py │ │ ├── test_monitor.py │ │ ├── test_vec │ │ │ ├── __init__.py │ │ │ ├── test_base.py │ │ │ ├── test_dummy.py │ │ │ └── test_subproc.py │ │ ├── test_wrap.py │ │ └── utils.py │ ├── test_nets.py │ ├── test_prob.py │ ├── test_sche.py │ └── test_utils.py └── utils.py ├── tools ├── print_scores.py └── run_unittest.sh └── unstable_baselines ├── __init__.py ├── algo ├── __init__.py ├── d │ ├── README.md │ ├── __init__.py │ ├── c51 │ │ ├── README.md │ │ ├── model.py │ │ ├── run.py │ │ └── train.sh │ ├── iqn │ │ ├── README.md │ │ ├── model.py │ │ ├── run.py │ │ └── train.sh │ └── qrdqn │ │ ├── README.md │ │ ├── model.py │ │ ├── run.py │ │ └── train.sh ├── dqn │ ├── README.md │ ├── __init__.py │ └── model.py ├── ppo │ ├── README.md │ ├── __init__.py │ └── model.py ├── s │ ├── README.md │ ├── __init__.py │ └── sac │ │ ├── README.md │ │ ├── model.py │ │ ├── run.py │ │ └── train.sh ├── sd3 │ ├── README.md │ ├── model.py │ ├── run.py │ └── train.sh └── td3 │ ├── README.md │ ├── model.py │ ├── run.py │ └── train.sh ├── dep ├── __init__.py ├── envs.py └── ppo │ ├── __init__.py │ ├── model_old.py │ └── run_old.py └── lib ├── __init__.py ├── base.py ├── data.py ├── envs ├── __init__.py ├── monitor.py ├── vec │ ├── __init__.py │ ├── base.py │ ├── dummy.py │ └── subproc.py └── wrap │ ├── __init__.py │ ├── atari.py │ └── common.py ├── nets.py ├── patch.py ├── prob.py ├── sche.py └── utils.py /.coveragerc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/.coveragerc -------------------------------------------------------------------------------- /.github/workflows/coverage.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/.github/workflows/coverage.yml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/.gitignore -------------------------------------------------------------------------------- /.gitmodules: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/.gitmodules -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/README.md -------------------------------------------------------------------------------- /example/README.md: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /example/atari/README.md: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /example/atari/train_dqn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/example/atari/train_dqn.py -------------------------------------------------------------------------------- /example/atari/train_ppo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/example/atari/train_ppo.py -------------------------------------------------------------------------------- /example/benchmark.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/example/benchmark.md -------------------------------------------------------------------------------- /example/cartpole/train_dqn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/example/cartpole/train_dqn.py -------------------------------------------------------------------------------- /example/cartpole/train_ppo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/example/cartpole/train_ppo.py -------------------------------------------------------------------------------- /example/pybullet/README.md: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/__init__.py: -------------------------------------------------------------------------------- 1 | from .base import BaseExperiments -------------------------------------------------------------------------------- /experiments/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/experiments/base.py -------------------------------------------------------------------------------- /experiments/ppo/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/experiments/ppo/__init__.py -------------------------------------------------------------------------------- /experiments/ppo/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/experiments/ppo/config.py -------------------------------------------------------------------------------- /experiments/ppo/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/experiments/ppo/run.py -------------------------------------------------------------------------------- /setup.cfg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/setup.cfg -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/setup.py -------------------------------------------------------------------------------- /test/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /test/test_algo/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /test/test_algo/test_dqn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_algo/test_dqn.py -------------------------------------------------------------------------------- /test/test_algo/test_ppo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_algo/test_ppo.py -------------------------------------------------------------------------------- /test/test_lib/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /test/test_lib/test_base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_base.py -------------------------------------------------------------------------------- /test/test_lib/test_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_data.py -------------------------------------------------------------------------------- /test/test_lib/test_envs/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /test/test_lib/test_envs/test_atari.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_envs/test_atari.py -------------------------------------------------------------------------------- /test/test_lib/test_envs/test_monitor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_envs/test_monitor.py -------------------------------------------------------------------------------- /test/test_lib/test_envs/test_vec/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /test/test_lib/test_envs/test_vec/test_base.py: -------------------------------------------------------------------------------- 1 | # We test the base and dummy together. -------------------------------------------------------------------------------- /test/test_lib/test_envs/test_vec/test_dummy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_envs/test_vec/test_dummy.py -------------------------------------------------------------------------------- /test/test_lib/test_envs/test_vec/test_subproc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_envs/test_vec/test_subproc.py -------------------------------------------------------------------------------- /test/test_lib/test_envs/test_wrap.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_envs/test_wrap.py -------------------------------------------------------------------------------- /test/test_lib/test_envs/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_envs/utils.py -------------------------------------------------------------------------------- /test/test_lib/test_nets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_nets.py -------------------------------------------------------------------------------- /test/test_lib/test_prob.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_prob.py -------------------------------------------------------------------------------- /test/test_lib/test_sche.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_sche.py -------------------------------------------------------------------------------- /test/test_lib/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/test_lib/test_utils.py -------------------------------------------------------------------------------- /test/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/test/utils.py -------------------------------------------------------------------------------- /tools/print_scores.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/tools/print_scores.py -------------------------------------------------------------------------------- /tools/run_unittest.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/tools/run_unittest.sh -------------------------------------------------------------------------------- /unstable_baselines/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/__init__.py -------------------------------------------------------------------------------- /unstable_baselines/algo/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /unstable_baselines/algo/d/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/d/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /unstable_baselines/algo/d/c51/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/c51/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/d/c51/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/c51/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/d/c51/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/c51/run.py -------------------------------------------------------------------------------- /unstable_baselines/algo/d/c51/train.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/c51/train.sh -------------------------------------------------------------------------------- /unstable_baselines/algo/d/iqn/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/iqn/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/d/iqn/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/iqn/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/d/iqn/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/iqn/run.py -------------------------------------------------------------------------------- /unstable_baselines/algo/d/iqn/train.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/iqn/train.sh -------------------------------------------------------------------------------- /unstable_baselines/algo/d/qrdqn/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/qrdqn/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/d/qrdqn/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/qrdqn/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/d/qrdqn/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/qrdqn/run.py -------------------------------------------------------------------------------- /unstable_baselines/algo/d/qrdqn/train.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/d/qrdqn/train.sh -------------------------------------------------------------------------------- /unstable_baselines/algo/dqn/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/dqn/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/dqn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/dqn/__init__.py -------------------------------------------------------------------------------- /unstable_baselines/algo/dqn/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/dqn/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/ppo/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/ppo/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/ppo/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/ppo/__init__.py -------------------------------------------------------------------------------- /unstable_baselines/algo/ppo/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/ppo/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/s/README.md: -------------------------------------------------------------------------------- 1 | # Soft RL 2 | 3 | ## Implemented Algorithms 4 | 5 | -------------------------------------------------------------------------------- /unstable_baselines/algo/s/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /unstable_baselines/algo/s/sac/README.md: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /unstable_baselines/algo/s/sac/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/s/sac/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/s/sac/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/s/sac/run.py -------------------------------------------------------------------------------- /unstable_baselines/algo/s/sac/train.sh: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /unstable_baselines/algo/sd3/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/sd3/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/sd3/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/sd3/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/sd3/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/sd3/run.py -------------------------------------------------------------------------------- /unstable_baselines/algo/sd3/train.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/sd3/train.sh -------------------------------------------------------------------------------- /unstable_baselines/algo/td3/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/td3/README.md -------------------------------------------------------------------------------- /unstable_baselines/algo/td3/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/td3/model.py -------------------------------------------------------------------------------- /unstable_baselines/algo/td3/run.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/td3/run.py -------------------------------------------------------------------------------- /unstable_baselines/algo/td3/train.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/algo/td3/train.sh -------------------------------------------------------------------------------- /unstable_baselines/dep/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /unstable_baselines/dep/envs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/dep/envs.py -------------------------------------------------------------------------------- /unstable_baselines/dep/ppo/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /unstable_baselines/dep/ppo/model_old.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/dep/ppo/model_old.py -------------------------------------------------------------------------------- /unstable_baselines/dep/ppo/run_old.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/dep/ppo/run_old.py -------------------------------------------------------------------------------- /unstable_baselines/lib/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/__init__.py -------------------------------------------------------------------------------- /unstable_baselines/lib/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/base.py -------------------------------------------------------------------------------- /unstable_baselines/lib/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/data.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/__init__.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/monitor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/monitor.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/vec/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/vec/__init__.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/vec/base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/vec/base.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/vec/dummy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/vec/dummy.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/vec/subproc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/vec/subproc.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/wrap/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/wrap/__init__.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/wrap/atari.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/wrap/atari.py -------------------------------------------------------------------------------- /unstable_baselines/lib/envs/wrap/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/envs/wrap/common.py -------------------------------------------------------------------------------- /unstable_baselines/lib/nets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/nets.py -------------------------------------------------------------------------------- /unstable_baselines/lib/patch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/patch.py -------------------------------------------------------------------------------- /unstable_baselines/lib/prob.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/prob.py -------------------------------------------------------------------------------- /unstable_baselines/lib/sche.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/sche.py -------------------------------------------------------------------------------- /unstable_baselines/lib/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Ending2015a/unstable_baselines/HEAD/unstable_baselines/lib/utils.py --------------------------------------------------------------------------------