search

katerakelly / oyster

Implementation of Efficient Off-policy Meta-learning via Probabilistic Context Variables (PEARL)
MIT License
473 stars 125 forks source link

RL2 benchmark #23

Closed janbolle closed 3 years ago

janbolle commented 3 years ago

Dear @katerakelly and @azhou42,

Thank you for providing your code for your exciting paper!

You are stating:

We ran our ProMP, MAML-TRPO, and RL2 baselines in the reference ProMP repo

But I cannot find the RL2 baselines. Can you please give me a hint where to look?

I really appreciate any help you can provide.

janbolle commented 3 years ago

or any RL2 implementation?

katerakelly commented 3 years ago

In the ProMP repository, switch to branch full_code, then see the run_scripts folder for the RL2 scripts.