Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
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 setclasses([...]), then predict.
Model source
Source description
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.
Derived from (OpenCLIP arch , pretrained tag ). Copyright (c) 2021 OpenAI; (c) 2012-2021 OpenCLIP authors. Licensed under the MIT License.
Sources
1 sourceVerified Sep 3
Model artifacts
1 artifactSource excerpts
2 excerpts
ViT-B-16laion2b_s34b_b88kThese 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.
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.
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))
MIT License. See the LICENSE and NOTICE files.
--- 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.
Source context: 0 downloads · 0 likes · Pipeline zero-shot-image-classification · Library libreyolo · Repo LibreYOLO/LibreCLIPb16-cls