In this third practical, you are asked to put what you just learnt
about function approximation. You are provided with the `main.py` file. Use `python main.py -h` to check how you are supposed to use this file.
In this project, you are asked to solve the classic Pendulum problem (https://gym.openai.com/envs/Pendulum-v0/).
Unlike previous environment, the state and action space are both continuous so that you need to approximate
the Q values Q(s, a). For more details about action and observation space, please refer to the OpenAI
documentation here: https://github.com/openai/gym/wiki/Pendulum-v0function approximation to good use.
For submission, you need to zip `agent.py` and `metadata` files then submit the zipped file to codalab.
`baseline.zip` as an example of submission.
If you want to reproduce your local score on Codalab, please use the docker image (https://cloud.docker.com/u/herilalaina/repository/docker/herilalaina/rlaic) and do not change the seed.
Then run `python main.py --ngames 1000 --niter 100 --batch 10`
For further questions, please use the codalab forum.
Submissions must be submitted before the 2018-12-17 01:42:00+00:00. You may submit 20 submissions every day and 10 in total.
Start: Jan. 7, 2019, midnight
Description: Development phase: create models and submit them or directly submit results on validation and/or test data; feed-back are provided on the validation set only.
Start: Jan. 18, 2019, 11 p.m.
Description: Final phase: submissions from the previous phase are automatically cloned and used to compute the final score. The results on the test set will be revealed when the organizers make them available.
Jan. 18, 2019, 11 p.m.
You must be logged in to participate in competitions.Sign In