Downloads · 30 days
0
oulianov/universr-lora-finetune-v1
universr-lora-finetune-v1 is a machine learning model from oulianov. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for pytorch.
An experimental LoRA adapter for the original UniverSR audio model. It is intended for audio bandwidth extension using the UniverSR pipeline.
Downloads · 30 days
0
Access
Public
Updated Jul 15, 2026
Repo size
34.2 MB
Likes
1
Public
Click a slice to open those files.
.pt34.2 MB · 100%
From the Hugging Face model README
An experimental LoRA adapter for the original UniverSR audio model. It is intended for audio bandwidth extension using the UniverSR pipeline.
adapter_model.pt is the final checkpoint from a rank-64 LoRA run trained from the original woongzip1/universr-audio weights. The run completed 720 epochs (51,120 optimizer steps) with a learning rate of 2e-4.
This repository contains the adapter only; load it onto the original UniverSR audio base model. The adapter metadata stores the rank, alpha, target modules, and other settings required to apply it.
The LoRA implementation and the UniverSR inference code are available in oulianov/UniverSR.