mirror of
https://github.com/wassname/ray.git
synced 2026-07-16 11:21:10 +08:00
SvenandEric Liang
f1b56fa5ee
* Unifying the code for PGTrainer/Policy wrt tf vs torch. Adding loss function test cases for the PGAgent (confirm equivalence of tf and torch). * Fix LINT line-len errors. * Fix LINT errors. * Fix `tf_pg_policy` imports (formerly: `pg_policy`). * Rename tf_pg_... into pg_tf_... following <alg>_<framework>_... convention, where ...=policy/loss/agent/trainer. Retire `PGAgent` class (use PGTrainer instead). * - Move PG test into agents/pg/tests directory. - All test cases will be located near the classes that are tested and then built into the Bazel/Travis test suite. * Moved post_process_advantages into pg.py (from pg_tf_policy.py), b/c the function is not a tf-specific one. * Fix remaining import errors for agents/pg/... * Fix circular dependency in pg imports. * Add pg tests to Jenkins test suite.
32 lines
789 B
Python
32 lines
789 B
Python
from ray.rllib.agents.trainer import with_common_config
|
|
from ray.rllib.agents.trainer_template import build_trainer
|
|
from ray.rllib.agents.pg.pg_tf_policy import PGTFPolicy
|
|
|
|
# yapf: disable
|
|
# __sphinx_doc_begin__
|
|
DEFAULT_CONFIG = with_common_config({
|
|
# No remote workers by default.
|
|
"num_workers": 0,
|
|
# Learning rate.
|
|
"lr": 0.0004,
|
|
# Use PyTorch as framework?
|
|
"use_pytorch": False
|
|
})
|
|
# __sphinx_doc_end__
|
|
# yapf: enable
|
|
|
|
|
|
def get_policy_class(config):
|
|
if config["use_pytorch"]:
|
|
from ray.rllib.agents.pg.pg_torch_policy import PGTorchPolicy
|
|
return PGTorchPolicy
|
|
else:
|
|
return PGTFPolicy
|
|
|
|
|
|
PGTrainer = build_trainer(
|
|
name="PG",
|
|
default_config=DEFAULT_CONFIG,
|
|
default_policy=PGTFPolicy,
|
|
get_policy_class=get_policy_class)
|