hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-05 18:11:42 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[RLlib] Fix some missing f-strings and a f-string related bug in tf eager policy. (#24148)

Browse source
This commit is contained in:
Noon van der Silk 2022-04-25 10:25:28 +01:00 • committed by GitHub
parent 56bc90ca72
commit 3589c21924
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
4 changed files with 6 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
rllib/evaluation/rollout_worker.py
View file

@ -1522,7 +1522,7 @@ class RolloutWorker(ParallelIteratorWorker):
f"PolicyID '{pid}' was probably added on-the-fly (not"
" part of the static `multagent.policies` config) and"
" no PolicySpec objects found in the pickled policy "
"state. Will not add `{pid}`, but ignore it for now."
f"state. Will not add `{pid}`, but ignore it for now."
)
else:
self.add_policy(

2
rllib/evaluation/sampler.py
View file

@ -745,7 +745,7 @@ def _env_runner(
simple_image_viewer.imshow(rendered)
elif rendered not in [True, False, None]:
raise ValueError(
"The env's ({base_env}) `try_render()` method returned an"
f"The env's ({base_env}) `try_render()` method returned an"
" unsupported value! Make sure you either return a "
"uint8/w x h x 3 (RGB) image or handle rendering in a "
"window and then return `True`."

4
rllib/policy/eager_tf_policy.py
View file

@ -516,8 +516,8 @@ def build_eager_tf_policy(
_is_training=tf.constant(False),
)
if state_batches is not None:
for s in enumerate(state_batches):
input_dict["state_in_{i}"] = s
for i, s in enumerate(state_batches):
input_dict[f"state_in_{i}"] = s
if prev_action_batch is not None:
input_dict[SampleBatch.PREV_ACTIONS] = prev_action_batch
if prev_reward_batch is not None:

4
rllib/utils/pre_checks/env.py
View file

@ -260,7 +260,7 @@ def check_multiagent_environments(env: "MultiAgentEnv") -> None:
if not env.action_space_contains(sampled_action):
error = (
_not_contained_error("action_space_sample", "action")
+ "\n\n sampled_action {sampled_action}\n\n"
+ f"\n\n sampled_action {sampled_action}\n\n"
)
raise ValueError(error)
@ -277,7 +277,7 @@ def check_multiagent_environments(env: "MultiAgentEnv") -> None:
if not env.observation_space_contains(next_obs):
error = (
_not_contained_error("env.step(sampled_action)", "observation")
+ ":\n\n next_obs: {next_obs} \n\n sampled_obs: {sampled_obs}"
+ f":\n\n next_obs: {next_obs} \n\n sampled_obs: {sampled_obs}"
)
raise ValueError(error)