File size: 1,616 Bytes
503481e 4b6a197 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 | ---
pretty_name: PAARBench PushObj Training Data
tags:
- robotics
- world-models
- test-time-adaptation
---
# Optional training dataset
The PAARBench evaluation protocol does **not** require training trajectories. Its goal
segments are tracked under `data/pushobj_eval/`, and evaluation reads fixed
normalization metadata without opening the training dataset.
The training release is provided for researchers who want to reproduce base-model or
adapter training, or develop methods that explicitly use offline trajectories:
```bash
.venv/bin/python scripts/download_data.py
```
The downloader fetches the immutable benchmark release from the
[`ThomasWalker1/paarbench-data`](https://huggingface.co/datasets/ThomasWalker1/paarbench-data)
dataset repository into `data/pushobj_multishape/` and verifies it against
`DATASET_MANIFEST.json`.
## Contents
The release contains only the nonredundant splits used by the training pipeline:
| split | episodes | files | bytes |
|---|---:|---:|---:|
| `train` | 7,200 | 7,206 | 864,106,305 |
| `val` | 800 | 806 | 97,366,292 |
Each split contains `states.pth`, `velocities.pth`, `rel_actions.pth`,
`abs_actions.pth`, `seq_lengths.pkl`, `shapes.pkl`, and one MP4 observation stream per
episode under `obses/`. Redundant root-level copies and intermediate shards are not
part of the release.
The trajectories were generated in the PushObj simulator across the T, L, Z, and +
shapes. No license is asserted here beyond the rights of the repository owner; users
should review the dataset repository card and applicable upstream asset licenses
before redistribution.
|