---
license: apache-2.0
language:
- en
tags:
- bridge_data
- v2
- image
- language
---

This dataset contains [`BridgeData V1` and `BridgeData V2`](https://rail-berkeley.github.io/bridgedata/), which is orignally downloaded from [this tar.gz file](https://rail.eecs.berkeley.edu/datasets/bridge_release/data/demos_8_17.zip), and use the following scripts to preprocess the data:
- [Preprocess V1](https://github.com/Kiteretsu77/This_and_That_VDM/blob/main/curation_pipeline/match_dataset_v1.py)
- [Preprocess V2](https://github.com/Kiteretsu77/This_and_That_VDM/blob/main/curation_pipeline/match_dataset_v2.py)
- [Train/test split](https://github.com/Kiteretsu77/This_and_That_VDM/blob/main/scripts/train_test_split.py)

After the processing, we have the dataset with the following structure:
| Folder name | Trajectory number | Size |
| :---------: | :---------------: | :--: |
| bridge_data_v1 (train) | 11007 | 30GB |
| bridge_data_v2 (train) | 16527 | 62GB |
| bridge_data_v1_test | 1222 | 3.3GB |
| bridge_data_v2_test | 1836 | 6.9GB |