act_pusht_bs64_dec7_100k

The original tuned champion recipe at 100k steps: 154.4 avg_sum_imputed_reward, 56.4% success (n=5000).

Part of a 6-week robot-learning training block (https://github.com/JaHeRoth/robot-learning). Evaluated on gym-pusht at n_action_steps=16; avg_sum_imputed_reward imputes 0.95/step to horizon 300 after success.

Downloads last month
-
Safetensors
Model size
84M params
Tensor type
F32
·
Video Preview
loading