Written for Jens Pommerening at Sereact. The dataset read here is lerobot/ucsd_pick_and_place_dataset. It is not Sereact's data, and it is on this page because a public pick and place corpus, the closest readable thing to the bin picking Sereact does.
1,355 episodes and 67,750 frames, and every single episode is exactly 50 frames, 100% of them at that length. That is a recording window rather than a task boundary, which decides what the last frames of an episode mean. Three task strings cover the corpus and the most common one accounts for 1,148 of the 1,355 episodes, 84.7%.
Read from lerobot/ucsd_pick_and_place_dataset. Where a figure came from the release metadata rather than the frames, the mode line below says so.
| property | value |
|---|---|
| episodes | 1,355 |
| frames | 67,750 |
| frame rate | 5 fps |
| robot | unknown |
| action dimensions | 4 |
| cameras | 1 |
| distinct task strings | 3 |
| episode length, shortest | 50 frames |
| episode length, median | 50 frames |
| episode length, longest | 50 frames |
Measurement mode, frame level, every row read.
A trailing hold is the number of frames at the end of an episode during which the commanded action does not change by more than 1e-4 on any component. Seconds come from the frame rate each release declares. Every corpus below was read the same way by the same code.
| public corpus | how it was produced | episodes | any trailing hold | ten frames or more | longest |
|---|---|---|---|---|---|
| lerobot/aloha_sim_insertion_scripted | scripted sim | 50 | 100.0% | 86.0% | 21 frames , 0.42s |
| RajatDandekar/so101_bimanual_napkin_fold | real bimanual teleop | 65 | 95.4% | 75.4% | 130 frames , 4.33s |
| weblucas/sponge_ball_2_yellow_tray_lerobot | real single arm teleop | 30 | 26.7% | 6.7% | 21 frames , 0.70s |
| lerobot/aloha_sim_insertion_human | sim human teleop | 50 | 6.0% | 0.0% | 1 frames , 0.02s |
| lerobot/ucsd_pick_and_place_dataset | sim, fixed window | 1,355 | 0.1% | 0.0% | 2 frames , 0.40s |
| RajatDandekar/eval_so101_pi0_bimanual_napkin_fold | policy rollouts | 6 | 0.0% | 0.0% | 0 frames , 0.00s |
The ordering is the finding. A script stops commanding when it is done and the last rows repeat. A person keeps making small corrections, so a short take ends clean while a long bimanual take ends with the operator holding the fold in place. A policy rollout ends the moment the episode is cut. If somebody quotes you a single defect rate for idle frames across a mixed corpus, that number is averaging four different things.
Each corpus is a meta directory plus one or more parquet files. Read data/chunk-000/file-000.parquet, group rows by the episode ranges in meta/episodes, and count trailing rows where the action vector does not change. No credentials and no download beyond the frame data itself.
Thirty eight public robot datasets, one at a time. Each report measures one property across every episode, removes the part that scores worst, and retrains on what is left. The corpus figures and the cuts are computed from the releases. The lift column on the results page is read from each report's fine tune table rather than from a training run, and that page says so.
The one result measured on physical hardware is Stera-10M. Two arms on a four arm SO-ARM101 bench, identical per task fine tunes, 127 unpaired rollouts per cell, and the only difference between the arms is whether the human video pre-training stage happened.
Everything above is a public release. The same four steps run on your data, and the first three cost nothing.
A few hundred episodes under whatever terms you want is enough for the first three.