mjf-su/PhysicalAI-US-ADE
收藏Hugging Face2026-03-27 更新2026-03-29 收录
下载链接:
https://hf-mirror.com/datasets/mjf-su/PhysicalAI-US-ADE
下载链接
链接失效反馈官方服务:
资源简介:
---
pretty_name: PhysicalAI-US-ADE
language:
- en
license: mit
task_categories:
- text-generation
tags:
- autonomous-driving
- trajectory-prediction
- evaluation
- waypoint-prediction
- physical-ai
- nvidia
- us-driving
size_categories:
- 100K<n<1M
---
# PhysicalAI-US-ADE
## Dataset Summary
**PhysicalAI-US-ADE** contains per-sample evaluation outputs for autonomous driving waypoint prediction on the **US subset** of the **PhysicalAI NVIDIA dataset**.
This dataset stores inference-time predictions and evaluation statistics for models evaluated on the dataset, organized by model name at the top level. Each model directory contains sample-level records for that model’s predictions against ground truth.
The current release includes evaluation results for:
- `base-AV-VLA` — corresponding to the model [`mjf-su/base-AV-VLA`](https://huggingface.co/mjf-su/base-AV-VLA)
The underlying source dataset used for evaluation is:
- [`tom-jerry-123/Physical-AI-AV-US`](https://huggingface.co/datasets/tom-jerry-123/Physical-AI-AV-US)
## What this dataset contains
Each JSONL record corresponds to a single evaluated sample and includes:
- sample identifiers:
- `scene_id`
- `chunk_name`
- `sample_idx`
- `timestamp_us`
- model outputs:
- `prediction_waypoints`
- reference targets:
- `ground_truth_waypoints`
- evaluation metrics:
- `ADE`
- `ADE_normalized`
- `ADE_combined`
- parsing / validity flags:
- `waypoint_count_match`
- `parse_fail`
## Directory structure
At the top level, each directory corresponds to one evaluated model.
Example:
```text
PhysicalAI-US-ADE/
├── base-AV-VLA/
│ ├── shard_00000.jsonl
│ ├── shard_00001.jsonl
│ ├── ...
│ └── CoT-targets.jsonl
---
### 数据集元数据
- 数据集名称:PhysicalAI-US-ADE
- 语言:英语
- 许可证:MIT协议
- 任务类别:文本生成
- 标签:自动驾驶(autonomous driving)、轨迹预测(trajectory prediction)、模型评估(evaluation)、路点预测(waypoint prediction)、PhysicalAI(physical-ai)、英伟达(NVIDIA)、美国驾驶场景(us-driving)
- 样本规模:10万至100万样本
---
# PhysicalAI-US-ADE 数据集
## 数据集概述
**PhysicalAI-US-ADE** 收录了基于 **PhysicalAI NVIDIA 数据集(PhysicalAI NVIDIA dataset)** 的**美国区域子集**所开展的自动驾驶路点预测任务的逐样本评估结果。
本数据集按照模型名称在顶层进行组织,存储了在该数据集上开展评估的各类模型的推理阶段预测结果与评估统计指标;每个模型对应的目录中,均包含该模型针对真实标注(ground truth)所生成的预测结果的逐样本记录。
当前版本收录了以下模型的评估结果:
- `base-AV-VLA` — 对应模型 [`mjf-su/base-AV-VLA`](https://huggingface.co/mjf-su/base-AV-VLA)
本次评估所使用的原始源数据集为:
- [`tom-jerry-123/Physical-AI-AV-US`](https://huggingface.co/datasets/tom-jerry-123/Physical-AI-AV-US)
## 数据集内容说明
每条JSONL格式的记录对应一个已完成评估的样本,包含以下内容:
- 样本标识符:
- `scene_id`(场景ID)
- `chunk_name`(分块名称)
- `sample_idx`(样本索引)
- `timestamp_us`(微秒级时间戳)
- 模型输出:
- `prediction_waypoints`(预测路点)
- 参考标注:
- `ground_truth_waypoints`(真实路点)
- 评估指标:
- `ADE`(平均位移误差)
- `ADE_normalized`(归一化平均位移误差)
- `ADE_combined`(组合平均位移误差)
- 解析与有效性标记:
- `waypoint_count_match`(路点数量匹配标记)
- `parse_fail`(解析失败标记)
## 目录结构
数据集顶层的每个目录对应一个待评估的模型。
示例目录结构如下:
text
PhysicalAI-US-ADE/
├── base-AV-VLA/
│ ├── shard_00000.jsonl
│ ├── shard_00001.jsonl
│ ├── ...
│ └── CoT-targets.jsonl
提供机构:
mjf-su



