# 🟦 Simulation Full-Shot Training Data / 仿真 Full-Shot 训练集

[Dataset Card](https://huggingface.co/datasets/wzlllong/value-model-benchmark-training-data#simulation) · [简体中文](https://huggingface.co/datasets/wzlllong/value-model-benchmark-training-data/blob/main/README_zh.md#simulation) · [Download & Read](https://huggingface.co/datasets/wzlllong/value-model-benchmark-training-data#download-and-reading)

**15 tasks × 100 demonstrations per task = 1,500 episodes.** All episodes belong to the `train` split (`0:1500`).

Based on open-source [RoboDojo](https://huggingface.co/datasets/RoboDojo-Benchmark/RoboDojo) data reviewed by our team, with selected demonstrations recorded again.

| Content | Specification |
| --- | --- |
| Platform | ARX |
| Dataset format | LeRobot Dataset v3.0 |
| Synchronized time steps | 789,612 |
| Sampling rate | 25 FPS |
| Robot state | `observation.state`: `float32[14]` |
| Robot action | `action`: `float32[14]` |
| RGB cameras | High, left wrist, right wrist: 640 × 480 each |
| Camera videos | 4,500 · AV1 / H.264 |

This directory is an independent LeRobot dataset root. Set `root` to `data/simulation/full-shot` inside the downloaded snapshot. The [download and training examples](https://huggingface.co/datasets/wzlllong/value-model-benchmark-training-data#download-and-reading) use LeRobot **0.4.2** and PyAV **15.1.0** and return states, actions, task instructions and three camera observations as PyTorch batches. Value or progress supervision is constructed by the training method.

[Feature definitions](https://huggingface.co/datasets/wzlllong/value-model-benchmark-training-data/blob/main/data/simulation/full-shot/meta/info.json) · [Shared trajectory contents and task descriptions](https://huggingface.co/datasets/wzlllong/value-model-benchmark-training-data#simulation)

<details>
<summary><strong>Per-task episode allocation</strong></summary>

| Task index | Source task key | Episodes | Episode range / index | Data-row range [start, end) |
| ---: | --- | ---: | --- | --- |
| 0 | `arrange_largest_number` | 100 | [0, 100) | [0, 58332) |
| 1 | `fill_pen_holder` | 100 | [100, 200) | [58332, 128073) |
| 2 | `fold_clothes` | 100 | [200, 300) | [128073, 158424) |
| 3 | `hang_mugs` | 100 | [300, 400) | [158424, 208157) |
| 4 | `insert_tubes` | 100 | [400, 500) | [208157, 240395) |
| 5 | `make_toast` | 100 | [500, 600) | [240395, 299974) |
| 6 | `organize_table` | 100 | [600, 700) | [299974, 355130) |
| 7 | `pack_objects_into_box` | 100 | [700, 800) | [355130, 436549) |
| 8 | `play_stacking_toy` | 100 | [800, 900) | [436549, 529154) |
| 9 | `press_by_number` | 100 | [900, 1000) | [529154, 568074) |
| 10 | `put_bottles_into_dustbin` | 100 | [1000, 1100) | [568074, 609854) |
| 11 | `stack_blocks` | 100 | [1100, 1200) | [609854, 636404) |
| 12 | `store_laptop_and_headphones` | 100 | [1200, 1300) | [636404, 688851) |
| 13 | `swap_blocks` | 100 | [1300, 1400) | [688851, 732459) |
| 14 | `sweep_blocks` | 100 | [1400, 1500) | [732459, 789612) |

</details>

Task indices match `meta/tasks.parquet`. Source episode identifiers and task correspondence are recorded in `provenance/`. `checksums.sha256` provides file integrity hashes.

**Publication authorization:** All data were authorized before public release.

## 中文

**15 个任务 × 每任务 100 条 = 1,500 条轨迹**，共 **789,612 个同步时间步**，全部属于 `train` 划分（`0:1500`）。

基于 [RoboDojo](https://huggingface.co/datasets/RoboDojo-Benchmark/RoboDojo) 开源数据整理，经团队审核后，对部分示例重新采集。

平台为 **ARX**，采用 **LeRobot Dataset v3.0**，采样率为 **25 FPS**。状态为 `float32[14]`，动作为 `float32[14]`；三路 RGB 相机为高位、左腕、右腕：均为 640 × 480，共 **4,500 个原编码视频（AV1 / H.264）**。

下载后，将 LeRobot 的 `root` 设置为快照中的 `data/simulation/full-shot`。主卡片的[下载与训练示例](https://huggingface.co/datasets/wzlllong/value-model-benchmark-training-data/blob/main/README_zh.md#download-and-reading)提供状态、动作、任务指令与三路相机观测的 PyTorch batch；value 或 progress 监督由训练方法构造。

折叠表列出任务索引、轨迹分配及数据行范围；原始来源标识与对应关系保留在 `provenance/`。

**发布授权：**全部数据均经授权后公开发布。