SARM complains about incorrect transformer version

[moured]

Ticket Type

🐛 Bug Report (Something isn't working)

Environment & System Info

LeRobot version: 0.4.4

Platform: Linux-6.8.0-51-generic-x86_64-with-glibc2.35

Python version: 3.10.12

Huggingface Hub version: 0.35.3

Datasets version: 4.1.1

Numpy version: 2.2.6

FFmpeg version: 4.4.2-0ubuntu0.22.04.1

PyTorch version: 2.7.1+cu126

Is PyTorch built with CUDA support?: True

Cuda version: 12.6

GPU model: NVIDIA H100 80GB HBM3

Using GPU in script?: <fill in>

lerobot scripts: ['lerobot-calibrate', 'lerobot-dataset-viz', 'lerobot-edit-dataset', 'lerobot-eval', 'lerobot-find-cameras', 'lerobot-find-joint-limits', 'lerobot-find-port', 'lerobot-imgtransform-viz', 'lerobot-info', 'lerobot-record', 'lerobot-replay', 'lerobot-setup-can', 'lerobot-setup-motors', 'lerobot-teleoperate', 'lerobot-train', 'lerobot-train-tokenizer']

Description

Hi,

I successfully trained a SARM reward model and generated the parquet file for the dataset However, when I start training pi05, it fails with the below error.

Note: I already did uv sync --extra sarm

Context & Reproduction

Training Command

accelerate launch \
  --multi_gpu \
  --num_processes 4 \
  --num_machines 1 \
  --mixed_precision bf16 \
  --dynamo_backend no \
  --main_process_port 0 \
  $(which lerobot-train) \
  --config_path=/path/to/train-policy.json \
  --use_rabc=true \
  --rabc_head_mode=sparse \
  --rabc_kappa=0.01

Relevant logs or stack trace

File "src/lerobot/policies/pi05/modeling_pi05.py", line 584, in __init__
raise ValueError(msg) from None
ValueError: An incorrect transformer version is used, please create an issue on https://github.com/huggingface/lerobot/issues

Checklist

• I have searched existing tickets to ensure this isn't a duplicate.
• I am using the latest version of the main branch.
• I have verified this is not an environment-specific problem.

Additional Info / Workarounds

I am not sure which transformer version could work. I tried different versions still got the same error. Any help?

[jadechoghari]

Run pip install -e ".[pi]

Pi models contain a special pinned version of transformers. This will be fixed in the future.