⚗️ experimenting training runs

2026-03-12 00:38:09 +01:00
parent 3b2d6d08f9
commit d3ed1c25ad
2 changed files with 50 additions and 19 deletions
--- a/configs/training/ppo_single.yaml
+++ b/configs/training/ppo_single.yaml
@@ -8,15 +8,17 @@ defaults:
  - _self_

 hidden_sizes: [256, 256]
-total_timesteps: 500000
-learning_epochs: 5
-learning_rate: 0.001
-entropy_loss_scale: 0.0001
-log_interval: 1024
+total_timesteps: 1000000
+learning_epochs: 10
+learning_rate: 0.0003
+entropy_loss_scale: 0.01
+rollout_steps: 2048
+mini_batches: 8
+log_interval: 2048
 checkpoint_interval: 10000
 initial_log_std: -0.5
 min_log_std: -4.0
-max_log_std: 0.0
+max_log_std: 2.0

 record_video_every: 50000