---
license: cc-by-4.0
pipeline_tag: image-to-image
tags:
- pytorch
- super-resolution
---

[Link to Github Release]()

# 4xNomos2_hq_atd
Scale: 4  
Architecture: [ATD](https://github.com/LabShuHangGU/Adaptive-Token-Dictionary)  
Architecture Option: [atd](https://github.com/muslll/neosr/blob/dc4e3742132bae2c2aa8e8d16de3a9fcec6b1a74/neosr/archs/atd_arch.py#L891)  

Author: Philip Hofmann  
License: CC-BY-0.4  
Purpose: Upscaler  
Subject: Photography  
Input Type: Images  
Release Date: 05.09.2024  

Dataset: [nomosv2](https://github.com/muslll/neosr/?tab=readme-ov-file#-datasets)  
Dataset Size: 6000  
OTF (on the fly augmentations): No  
Pretrained Model: 003_ATD_SRx4_finetune  
Iterations: 180'000  
Batch Size: 2  
Patch Size: 48  
Norm: true  

Description:   
An atd 4x upscaling model, similiar to the [4xNomos2_hq_dat2](https://github.com/Phhofm/models/releases/tag/4xNomos2_hq_dat2) or [4xNomos2_hq_mosr](https://github.com/Phhofm/models/releases/tag/4xNomos2_hq_mosr) models, trained and for usage on non-degraded input to give good quality output. 

Training checkpoints metric scoring on val images  
![image](https://github.com/user-attachments/assets/2ae3b750-adbd-4ceb-8ff3-cafee2ee97d1)  

Showcase of the top 3 checkpoints from this model training, where 180k has been selected as the main release model: https://slow.pics/c/ZEnoG0Ou  
I added the other checkpoints (135k and 205k) as additional model files in the assets of this release.  

## Model Showcase:
[Slowpics](https://slow.pics/c/ttYvxmJq)

(Click on image for better view)
![Example1](https://github.com/user-attachments/assets/b4ec00fb-464c-4039-b50f-656c8a49a2b4)
![Example2](https://github.com/user-attachments/assets/3cd16733-f14d-424a-94e5-ef812aa9a1dd)
![Example3](https://github.com/user-attachments/assets/4d19f3e5-5844-437f-9805-9a1eb50d79db)
![Example4](https://github.com/user-attachments/assets/c7246289-faed-4f26-a987-da1f88eb6f33)
![Example5](https://github.com/user-attachments/assets/09286ce5-6f64-467a-a9db-a6ddbc803b32)