hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-06 02:21:39 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[RLlib] Update sac_tf_policy.py (add tf.cast to float32 for rewards) (#14843)

Browse source
This commit is contained in:
astronauti 2021-03-24 16:12:55 +01:00 • committed by GitHub
parent 6708211b59
commit 8874ccec2d
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
rllib/agents/sac/sac_tf_policy.py
View file

@ -323,7 +323,7 @@ def sac_actor_critic_loss(
# Compute RHS of bellman equation for the Q-loss (critic(s)).
q_t_selected_target = tf.stop_gradient(
train_batch[SampleBatch.REWARDS] +
tf.cast(train_batch[SampleBatch.REWARDS], tf.float32) +
policy.config["gamma"]**policy.config["n_step"] * q_tp1_best_masked)
# Compute the TD-error (potentially clipped).