Instructions to use thepatch/same-l-decoder-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Stable Audio 3
How to use thepatch/same-l-decoder-lora with Stable Audio 3:
# 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
File size: 1,192 Bytes
3b32f6c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 | This Stability AI Model is licensed under the Stability AI Community License,
Copyright © Stability AI Ltd. All Rights Reserved
ATTRIBUTION FOR THE STABILITY AI MATERIALS
The notice above applies to the Stability AI Materials this work is derived from:
SAME-L, the autoencoder published at https://huggingface.co/stabilityai/SAME-L.
A copy of the Stability AI Community License Agreement is included in this
repository as LICENSE.md.
WHAT THIS DERIVATIVE WORK CHANGES, AND HOW
squeakfix_v3.safetensors is a Derivative Work of SAME-L, trained by the collabage
patch (https://huggingface.co/thepatch).
It is a rank-16 low-rank adaptation of SAME-L's decoder. LoRA adapters are
attached to 98 nn.Linear modules of the decoder, about 1.30 % of the decoder's
parameters, and trained with a self-supervised objective that adds round-trip
consistency to the decoder's rendering of latents.
SAME-L itself is not modified or redistributed here. The encoder is untouched and
none of SAME-L's own weights are included in this repository; the adapter is a
separate file applied to SAME-L at load time. The audio in samples/ was produced
by decoding through SAME-L with and without this adapter.
|