README.md

---
license: cc-by-nc-4.0
paperswithcode_id: vggface2
pretty_name: vggface2
---

## Bounded streaming

`datasets==5.0.0` does not execute this repository's remote Python loader
through `load_dataset()`. Clone code and metadata without downloading Git LFS
objects, then use the project-side module directly:

```shell
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/datasets/ProgramComputer/VGGFace2
```

Archive bytes are read sequentially and are not extracted or cached.

```python
from VGGFace2 import load_streaming

dataset = load_streaming(
    split="train",
    revision="ad5f6b5a5f560621fd7efb9b79c956d27d427a08",
    cache_dir="./metadata-cache",
    scratch_dir="./stream-scratch",
)
first_hundred = list(dataset.take(100))
```

```
@article{DBLP:journals/corr/abs-1710-08092,
  author       = {Qiong Cao and
                  Li Shen and
                  Weidi Xie and
                  Omkar M. Parkhi and
                  Andrew Zisserman},
  title        = {VGGFace2: {A} dataset for recognising faces across pose and age},
  journal      = {CoRR},
  volume       = {abs/1710.08092},
  year         = {2017},
  url          = {http://arxiv.org/abs/1710.08092},
  eprinttype    = {arXiv},
  eprint       = {1710.08092},
  timestamp    = {Wed, 04 Aug 2021 07:50:14 +0200},
  biburl       = {https://dblp.org/rec/journals/corr/abs-1710-08092.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}
```