This is the pytorch version of the A2C - which is basiclly the same as the openai baselines
- python-3.5.2
- openai baselines
- pytorch-0.4.0
Install OpenAI Baselines (the openai-baselines update so quickly, please use the older version as blow, will solve in the future.)
# clone the openai baselines
git clone https://github.com/openai/baselines.git
cd baselines
git checkout 366f486
pip install -e .
python train_network.py --env-name 'BreakoutNoFrameskip-v4' --cuda (if you have a GPU, you can add this flag)
python demo.py --env-name 'BreakoutNoFrameskip-v4'
Please download them from the Google Driver, then put the saved_models under the current folder.
It can be found some learning curves was not perfect, because it's the problem of the implementation of RMSProp in pytorch. Please check this blog
