Downloads · 30 days
0
Linorman616/rnapy_models
rnapy_models is a machine learning model from Linorman616. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This repository aggregates pretrained weights from several open-source RNA-related deep learning models for convenient access and hosting on Hugging Face. No code from the original projects is redistributed here—only…
Downloads · 30 days
0
Access
Public
Updated Sep 22, 2025
Repo size
5.3 GB
Likes
0
Public
Click a slice to open those files.
.pth4.4 GB · 83%
From the Hugging Face model README
This repository aggregates pretrained weights from several open-source RNA-related deep learning models for convenient access and hosting on Hugging Face. No code from the original projects is redistributed here—only the checkpoint files as released (or lightly renamed) by the original authors.
IMPORTANT: You must consult and comply with the original licenses and citation requirements of each upstream project before using these weights in research or downstream applications.
| Model | File(s) in models/ | Original Repository | Task / Summary |
|---|---|---|---|
| RNA-FM | rna_fm_pretrained.pth, mrna_fm_pretrained.pth | https://github.com/ml4bio/RNA-FM | Foundation model for RNA sequences (masked language modeling over nucleotides). |
| RhoFold | rho_fold_pretrained.pt | https://github.com/ml4bio/RhoFold | End-to-end RNA 3D structure prediction (folding) leveraging pretrained sequence embeddings. |
| RiboDiffusion | ribo_diffusion_pretrained.pth, ribo_diffusion_pretrained_large.pth | https://github.com/ml4bio/RiboDiffusion | Diffusion-based generative modeling for RNA (e.g. inverse design / structure generation). |
| RhoDesign | rho_design_pretrained.pth | https://github.com/ml4bio/RhoDesign | RNA sequence design conditioned on secondary / tertiary constraints using pretrained embeddings. |
| RNA-MSM | rna_msm_pretrained.pt | https://github.com/yikunpku/RNA-MSM | Multi-scale RNA structural representation model (MSA + pairwise + 3D context). |
If a model provides multiple files, they correspond to alternative serialization formats or size variants supplied upstream (e.g., small / large checkpoints).
models/
mrna_fm_pretrained.pth
rho_design_pretrained.pth
rho_design_pretrained_large.pth
rho_fold_pretrained.pt
ribo_diffusion_pretrained.pth
rna_fm_pretrained.pth
rna_msm_pretrained.pt
README.md