hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-05 10:01:43 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[RLlib]: Fix OPE docs. (#27460)

Browse source
This commit is contained in:
Rohan Potdar 2022-08-11 12:44:22 +05:30 • committed by GitHub
parent 894e19f791
commit 600b8d4729
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
doc/source/rllib/rllib-offline.rst
View file

@ -124,11 +124,11 @@ We can now train a DQN algorithm offline and evaluate it using OPE:
"wis": {"type": WeightedImportanceSampling},
"dm_fqe": {
"type": DirectMethod,
"q_model_config": {"type": FQETorchModel, "tau": 0.05},
"q_model_config": {"type": FQETorchModel, "polyak_coef": 0.05},
},
"dr_fqe": {
"type": DoublyRobust,
"q_model_config": {"type": FQETorchModel, "tau": 0.05},
"q_model_config": {"type": FQETorchModel, "polyak_coef": 0.05},
},
},
)