---
license: apache-2.0
library_name: libreyolo
pipeline_tag: zero-shot-image-classification
tags:
  - libreyolo
  - perception-encoder
  - zero-shot-image-classification
  - image-feature-extraction
  - clip
---

# LibrePEt16-cls

Perception Encoder (PE) Core **PE-Core-T/16-384** packaged for
[LibreYOLO](https://github.com/LibreYOLO/libreyolo).

PE Core is a dual-tower vision-language encoder: it maps images, text, and whole
videos into one normalized embedding space, and classifies zero-shot from text
prompts.

```python
from libreyolo import LibreYOLO

# zero-shot classification
model = LibreYOLO("LibrePEt16-cls.pt")
model.set_classes(["a forklift", "an empty aisle", "a spill"])
r = model.predict("warehouse.jpg")[0]
print(model.names[r.probs.top1], float(r.probs.top1conf))

# embeddings: image, text, and whole video in one space
embedder = LibreYOLO("LibrePEt16-cls.pt", task="embed")
embedder.predict("photo.jpg")      # (1, 512) image row
embedder.embed_text(["a dog"])     # (1, 512) text row
embedder.predict("clip.mp4")       # (1, 512) row for the whole clip
```

The bare canonical filename is the download trigger - there is no
`from_pretrained`.

## Specification

| | |
| --- | --- |
| Family / size | `pe` / `t16` |
| Tasks | `classify` (zero-shot), `embed` |
| Input resolution | 384x384 |
| Embedding dimension | 512 |
| Text context length | 32 |
| Preprocessing | RGB, bilinear square resize, mean/std = (0.5, 0.5, 0.5) |
| Video pooling | mean of independently encoded frames, L2-normalized once |
| Training | not supported (inference-only; `train()` raises) |
| Exports | ONNX, TorchScript |

## Source

- Upstream repository: [`timm/PE-Core-T-16-384`](https://huggingface.co/timm/PE-Core-T-16-384)
- Upstream revision: `7fe539ed578ac49a1c2b4f946e4b0747704c825a`
- Upstream license: Apache-2.0

This is a converted **OpenCLIP-compatible** PE Core checkpoint, not an
unmodified official `facebook/PE-Core-*` package checkpoint.

## Modifications

State-dict metadata wrap only (`weights/convert_pe_weights.py`): LibreYOLO
checkpoint metadata is added and tensors are moved to CPU. **Learned parameters
are unchanged.**

Parity against unmodified `open_clip_torch==3.2.0` on float32 CPU is exact
(`max_abs_diff == 0.0`) for image embeddings, text embeddings, zero-shot logits,
and fixed-frame video embeddings.

## License

Apache-2.0, inherited from the upstream weights. See `LICENSE` and `NOTICE` in
this repository. You are responsible for complying with those terms.