Instructions to use LibreYOLO/LibreMobileSAM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LibreYOLO
How to use LibreYOLO/LibreMobileSAM with LibreYOLO:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Download NOTICE from LibreYOLO/LibreMobileSAM: direct link, hf CLI and curl.
- Browser
- Download file 997 Bytes
-
https://huggingface.co/LibreYOLO/LibreMobileSAM/resolve/main/NOTICE
- Command line
-
hf download hf://LibreYOLO/LibreMobileSAM/NOTICE
-
curl -L -o NOTICE https://huggingface.co/LibreYOLO/LibreMobileSAM/resolve/main/NOTICE
997 Bytes
| LibreMobileSAM - third-party attribution | |
| ======================================== | |
| The MobileSAM architecture in this family is a native LibreYOLO port derived | |
| from: | |
| Project: MobileSAM | |
| Upstream: https://github.com/ChaoningZhang/MobileSAM | |
| License: Apache License 2.0 | |
| The TinyViT image encoder portions preserve the upstream TinyViT module and | |
| state-dict structure: | |
| Project: TinyViT | |
| Authors: Microsoft contributors | |
| License: Apache License 2.0 | |
| The prompt encoder, two-way transformer, mask decoder, preprocessing geometry, | |
| and mask postprocessing follow the MobileSAM v1 implementation so the converted | |
| checkpoint loads with strict key matching and can be checked for bit-exact | |
| parity against the upstream Apache-2.0 implementation. | |
| Pretrained weights are converted from the MobileSAM Apache-2.0 checkpoint and | |
| hosted as LibreMobileSAM.pt under LibreYOLO/LibreMobileSAM. The hosted model | |
| repo must carry this NOTICE and the Apache-2.0 LICENSE alongside the weights. | |