System Info

- lerobot version: 0.4.0
- Platform: Linux-6.8.0-86-generic-x86_64-with-glibc2.35
- Python version: 3.11.13
- Huggingface Hub version: 0.35.3
- Datasets version: 4.1.1
- Numpy version: 2.3.4
- PyTorch version: 2.7.1+cu126
- Is PyTorch built with CUDA support?: True
- Cuda version: 12.6
- GPU model: Quadro M2000
- Using GPU in script?: no

Information

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

Reproduction

Set video_encoding_batch_size to a value larger than 1 (e.g. 4) and run the recording script:

lerobot-record
--dataset.video_encoding_batch_size=4
--robot.type=so101_follower
--robot.port=/dev/tty.usbmodem585A0076841
--robot.id=my_awesome_follower_arm
--robot.cameras="{ front: {type: opencv, index_or_path: 0, width: 1920, height: 1080, fps: 30}}"
--teleop.type=so101_leader
--teleop.port=/dev/tty.usbmodem58760431551
--teleop.id=my_awesome_leader_arm
--display_data=true
--dataset.repo_id=${HF_USER}/record-test
--dataset.num_episodes=5
--dataset.single_task="Grab the black cube"

I got the following error:

INFO 2025-11-07 09:54:26 llo/gello.py:79 gello_teleop Gello connected.
INFO 2025-11-07 09:54:26 ls/utils.py:227 Recording episode 0
Right arrow key pressed. Exiting loop...
^[[CINFO 2025-11-07 09:54:33 ls/utils.py:227 Reset the environment
Right arrow key pressed. Exiting loop...
^[[CRight arrow key pressed. Exiting loop...
Map: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 175/175 [00:00<00:00, 1294.68 examples/s]
INFO 2025-11-07 09:54:40 eo_utils.py:634 Exception occurred. Encoding remaining episodes before exit...
INFO 2025-11-07 09:54:40 eo_utils.py:640 Encoding remaining 1 episodes, from episode 0 to 0
INFO 2025-11-07 09:54:40 dataset.py:1182 Batch encoding 4 videos for episodes 0 to 0
Traceback (most recent call last):
File "/home/finn/dev/lerobot_ur5e_gello/scripts/record.py", line 447, in record
dataset.meta.episodes = load_episodes(dataset.meta.root)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/lerobot/datasets/utils.py", line 365, in load_episodes
episodes = load_nested_dataset(local_dir / EPISODES_DIR)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/lerobot/datasets/utils.py", line 121, in load_nested_dataset
datasets = Dataset.from_parquet([str(path) for path in paths], features=features)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/datasets/arrow_dataset.py", line 1303, in from_parquet
).read()
^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/datasets/io/parquet.py", line 61, in read
self.builder.download_and_prepare(
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/datasets/builder.py", line 894, in download_and_prepare
self._download_and_prepare(
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/datasets/builder.py", line 948, in _download_and_prepare
split_generators = self._split_generators(dl_manager, **split_generators_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/datasets/packaged_modules/parquet/parquet.py", line 60, in _split_generators
self.info.features = datasets.Features.from_arrow_schema(pq.read_schema(f))
^^^^^^^^^^^^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/pyarrow/parquet/core.py", line 2393, in read_schema
file = ParquetFile(
^^^^^^^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/pyarrow/parquet/core.py", line 328, in init
self.reader.open(
File "pyarrow/_parquet.pyx", line 1656, in pyarrow._parquet.ParquetReader.open
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
pyarrow.lib.ArrowInvalid: Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/home/finn/dev/lerobot_ur5e_gello/scripts/record.py", line 474, in
main()
File "/home/finn/dev/lerobot_ur5e_gello/scripts/record.py", line 470, in main
record()
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/lerobot/configs/parser.py", line 233, in wrapper_inner
response = fn(cfg, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/finn/dev/lerobot_ur5e_gello/scripts/record.py", line 393, in record
with VideoEncodingManager(dataset):
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/lerobot/datasets/video_utils.py", line 644, in exit
self.dataset._batch_save_episode_video(start_ep, end_ep)
File "/home/finn/dev/lerobot_ur5e_gello/.venv/lib/python3.11/site-packages/lerobot/datasets/lerobot_dataset.py", line 1186, in _batch_save_episode_video
chunk_idx = self.meta.episodes[start_episode]["data/chunk_index"]
~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^
TypeError: 'NoneType' object is not subscriptable

Expected behavior

The recording should work as usual but with batch encoding.