RLGC-Project / RLGC

An open-source platform for applying Reinforcement Learning for Grid Control (RLGC)
Other
108 stars 30 forks source link

about storedData #5

Closed diving16 closed 4 years ago

diving16 commented 4 years ago

When I run the train files like "trainKundur2areaGenBrakingAgent", "trainIEEE39LoadSheddingAgent_continuous_action_DDPG" or "trainIEEE39LoadSheddingAgent_discreteaction", I found generated npy.files in storedData is 1kB and all the step* variables like "step_rewards" become []. Thus I cannot check it as the .ipynb files you provided. I have already build the environment uses the yml environment.

thuang commented 4 years ago

There are some discrepancies as .ipynb files were not updated and some parts of the callback function in the training *.py files was commented.

You may follow the stable-baselines (different from OpenAI baselines) example below to monitor and check the training results.

https://stable-baselines.readthedocs.io/en/master/guide/examples.html#using-callback-monitoring-training