From 7dc9ffe4c9625b4eff4bdfa5eecb17ea47b2d9fe Mon Sep 17 00:00:00 2001 From: Huan Liu Date: Sat, 15 Mar 2025 00:07:14 +0800 Subject: [PATCH] Update 10_use_so100.md (#840) --- examples/10_use_so100.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/examples/10_use_so100.md b/examples/10_use_so100.md index b8b45aa5..d2423229 100644 --- a/examples/10_use_so100.md +++ b/examples/10_use_so100.md @@ -583,6 +583,13 @@ Let's explain it: Training should take several hours. You will find checkpoints in `outputs/train/act_so100_test/checkpoints`. +To resume training from a checkpoint, below is an example command to resume from `last` checkpoint of the `act_so100_test` policy: +```bash +python lerobot/scripts/train.py \ + --config_path=outputs/train/act_so100_test/checkpoints/last/pretrained_model/train_config.json \ + --resume=true +``` + ## K. Evaluate your policy You can use the `record` function from [`lerobot/scripts/control_robot.py`](../lerobot/scripts/control_robot.py) but with a policy checkpoint as input. For instance, run this command to record 10 evaluation episodes: