Initial upload: LibreCLIP zero-shot weights

LICENSE

@@ -0,0 +1,22 @@
+MIT License
+
+Copyright (c) 2021 OpenAI
+Copyright (c) 2012-2021 OpenCLIP authors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

LibreCLIPb16-cls.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a57c2f5bcca165091689c6b4171d0be9ae11d00999b64416a18f568ded66e2d8
+size 598618871

NOTICE

@@ -0,0 +1,16 @@
+LibreCLIP weights
+-----------------
+
+This product contains weights derived from OpenCLIP
+(https://github.com/mlfoundations/open_clip), built on OpenAI CLIP
+(https://github.com/openai/CLIP).
+Copyright (c) 2021 OpenAI; (c) 2012-2021 OpenCLIP authors.
+Licensed under the MIT License.
+
+Source weights: laion/CLIP-ViT-B-16-laion2B-s34B-b88K
+(OpenCLIP arch ViT-B-16, pretrained tag laion2b_s34b_b88k).
+
+Data provenance: trained on LAION-2B, which had a documented CSAM-content
+history (Stanford, December 2023). LAION subsequently released Re-LAION, a
+cleaned re-release. Prefer Re-LAION-derived weights where available. LibreCLIP
+does not ship OpenAI-WIT (undisclosed-data) or any non-commercial CLIP weights.

README.md

@@ -0,0 +1,51 @@
+---
+license: mit
+library_name: libreyolo
+pipeline_tag: zero-shot-image-classification
+tags:
+- clip
+- zero-shot-image-classification
+- open-vocabulary
+- libreyolo
+---
+
+# LibreCLIPb16-cls
+
+OpenCLIP ViT-B/16 (LAION-2B), repackaged as a native LibreYOLO checkpoint for
+**zero-shot, open-vocabulary** image classification with `LibreCLIP`. No
+training and no fixed label set: call `set_classes([...])`, then predict.
+
+## Source
+
+Derived from [laion/CLIP-ViT-B-16-laion2B-s34B-b88K](https://huggingface.co/laion/CLIP-ViT-B-16-laion2B-s34B-b88K)
+(OpenCLIP arch `ViT-B-16`, pretrained tag `laion2b_s34b_b88k`).
+Copyright (c) 2021 OpenAI; (c) 2012-2021 OpenCLIP authors. Licensed under the
+MIT License.
+
+## Data provenance
+
+These weights were trained on LAION-2B, which has a documented CSAM-content
+history (Stanford, December 2023); LAION subsequently released the cleaned
+Re-LAION. Prefer Re-LAION-derived weights where available. See `NOTICE`.
+
+## Modifications
+
+State-dict key remapping only — LibreCLIP's native towers mirror the OpenCLIP
+module structure, so the load is 0-missing / 0-unexpected. Learned parameters
+are unchanged. See `weights/convert_clip_weights.py` in the
+[LibreYOLO source repository](https://github.com/LibreYOLO/libreyolo).
+
+## Usage
+
+```python
+from libreyolo import LibreCLIP
+
+model = LibreCLIP("LibreCLIPb16-cls.pt")                 # autodownloads from this repo
+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))
+```
+
+## License
+
+MIT License. See the [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE) files.