Running a model trained earlier leads to stats error

[Himhawkins]

System Info

Error: File "/home/pranav/Repo/lerobot/lerobot/common/policies/normalize.py", line 169, in forward
    assert not torch.isinf(mean).any(), _no_stats_error_str("mean")
AssertionError: `mean` is infinity. You should either initialize with `stats` as an argument, or use a pretrained model.
 Can someone help with resolving this? I am getting this while running evaluate on an ACT model i trained. Model: ps5387/SHAKE29k

Information

• One of the scripts in the examples/ folder of LeRobot
• My own task or dataset (give details below)

Reproduction

Download the ps5387/SHAKE29k and running python lerobot/scripts/control_robot.py
--robot.type=so100
--control.type=record
--control.fps=30
--control.single_task="Grab and Shake hand when you see it"
--control.repo_id=ps5387/eval_act_so100_shake2_1
--control.tags='["so100","shake"]'
--control.warmup_time_s=5
--control.episode_time_s=25
--control.reset_time_s=10
--control.num_episodes=10
--control.push_to_hub=true
--control.policy.path=shake20k/pretrained_model

Expected behavior

Getting error File "/home/pranav/Repo/lerobot/lerobot/common/policies/normalize.py", line 169, in forward
assert not torch.isinf(mean).any(), _no_stats_error_str("mean")
AssertionError: mean is infinity. You should either initialize with stats as an argument, or use a pretrained model.

[CarolinePascal]

HI @Himhawkins !

Could you check if the naming issue reported in #1095 solves your issue as well ?

Best,

Caroline.