W1122 17:46:10.678000 81452 site-packages/torch/distributed/elastic/multiprocessing/redirects.py:35] NOTE: Redirects are currently not supported in MacOs. objc[81452]: Class AVFFrameReceiver is implemented in both /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/python3.12/site-packages/cv2/.dylibs/libavdevice.61.3.100.dylib (0x105da43a8) and /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/python3.12/site-packages/av/.dylibs/libavdevice.61.3.100.dylib (0x130d803a8). This may cause spurious casting failures and mysterious crashes. One of the duplicates must be removed or renamed. objc[81452]: Class AVFAudioReceiver is implemented in both /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/python3.12/site-packages/cv2/.dylibs/libavdevice.61.3.100.dylib (0x105da43f8) and /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/python3.12/site-packages/av/.dylibs/libavdevice.61.3.100.dylib (0x130d803f8). This may cause spurious casting failures and mysterious crashes. One of the duplicates must be removed or renamed. INFO 2026-11-22 17:46:14 t_record.py:401 {'dataset': {'depth_encoder': {'crf': 30, 'depth_max': 10.0, 'depth_min': 0.01, 'extra_options': {'x265-params': 'lossless=1'}, 'fast_decode': 0, 'g': 2, 'pix_fmt': 'gray12le', 'preset': None, 'shift': 3.5, 'use_log': True, 'vcodec': 'hevc', 'video_backend': 'pyav'}, 'encoder_queue_maxsize': 30, 'encoder_threads': None, 'episode_time_s': 10, 'fps': 30, 'no_stamp': True, 'num_episodes': 3, 'num_image_writer_processes': 0, 'num_image_writer_threads_per_camera': 4, 'private': None, 'push_to_hub': False, 'repo_id': 'local/check_258', 'reset_time_s': 10, 'rgb_encoder': {'crf': 30, 'extra_options': {}, 'fast_decode': 0, 'g': 2, 'pix_fmt': 'yuv420p', 'preset': 12, 'vcodec': 'libsvtav1', 'video_backend': 'pyav'}, 'root': '/Users/shimazueiichi/r2r2r-scripts/so101/check_258/dataset', 'single_task': 'check 258 practice', 'streaming_encoding': False, 'tags': None, 'video': True, 'video_encoding_batch_size': 1}, 'display_compressed_images': False, 'display_data': False, 'display_ip': None, 'display_mode': 'rerun', 'display_port': None, 'play_sounds': False, 'resume': False, 'robot': {'calibration_dir': None, 'cameras': {'desk': {'backend': , 'color_mode': , 'fourcc': 'MJPG', 'fps': 30, 'height': 480, 'index_or_path': 0, 'rotation': , 'warmup_s': 1, 'width': 640}, 'wrist': {'backend': , 'color_mode': , 'fourcc': 'MJPG', 'fps': 30, 'height': 480, 'index_or_path': 1, 'rotation': , 'warmup_s': 1, 'width': 640}}, 'disable_torque_on_disconnect': True, 'id': 'ptm_follower', 'max_relative_target': None, 'num_read_retries': 2, 'port': '/dev/tty.usbmodem5CBC4321631', 'position_d_coefficient': 32, 'position_i_coefficient': 0, 'position_p_coefficient': 16, 'use_degrees': True}, 'teleop': {'calibration_dir': None, 'id': 'ptm_leader', 'num_read_retries': 2, 'port': '/dev/tty.usbmodem5CBC4323581', 'use_degrees': True}} objc[81452]: Class AVFFrameReceiver is implemented in both /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/python3.12/site-packages/cv2/.dylibs/libavdevice.61.3.100.dylib (0x105da43a8) and /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/libavdevice.59.7.100.dylib (0x1592c8778). This may cause spurious casting failures and mysterious crashes. One of the duplicates must be removed or renamed. objc[81452]: Class AVFAudioReceiver is implemented in both /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/python3.12/site-packages/cv2/.dylibs/libavdevice.61.3.100.dylib (0x105da43f8) and /opt/homebrew/Caskroom/miniforge/base/envs/lerobot/lib/libavdevice.59.7.100.dylib (0x1592c87c8). This may cause spurious casting failures and mysterious crashes. One of the duplicates must be removed or renamed. INFO 2026-11-22 17:46:18 so_leader.py:95 ptm_leader SOLeader connected. WARNING 2026-11-22 17:46:18 a_opencv.py:284 OpenCVCamera(1) failed to set fourcc=MJPG (actual=, success=False). Continuing with default format. INFO 2026-11-22 17:46:20 a_opencv.py:198 OpenCVCamera(1) connected. WARNING 2026-11-22 17:46:21 a_opencv.py:284 OpenCVCamera(0) failed to set fourcc=MJPG (actual=, success=False). Continuing with default format. INFO 2026-11-22 17:46:23 a_opencv.py:198 OpenCVCamera(0) connected. INFO 2026-11-22 17:46:23 follower.py:128 ptm_follower SOFollower connected. INFO 2026-11-22 17:46:23 rd_input.py:375 Keyboard listener started (Right/Left/Esc, or n=next, r=re-record, q=quit). INFO 2026-11-22 17:46:23 t_record.py:504 Streaming encoding is disabled. If you have capable hardware, consider enabling it for way faster episode saving. --dataset.streaming_encoding=true --dataset.encoder_threads=2 # --dataset.rgb_encoder.vcodec=auto. More info in the documentation: https://huggingface.co/docs/lerobot/streaming_video_encoding INFO 2026-11-22 17:46:23 ls/utils.py:151 Recording episode 0 INFO 2026-11-22 17:46:33 ls/utils.py:151 Reset the environment Map: 0%| | 0/300 [00:00