Edit model card

Link to Github Release

2xHFA2kCompact

Name: 2xHFA2kCompact
Author: Philip Hofmann
Release Date: 18.04.2023
License: CC BY 4.0
Network: SRVGGNetCompact, with parameters: 600,652
Scale: 2
Purpose: A compact anime 2x upscaling model based on musl's HFA2k dataset

Iterations: 93,000
batch_size: 12
HR_size: 384
Epoch: 207 (require iter number per epoch: 214)
Dataset: HFA2k
Number of train images: 2568
OTF Training: Yes
Pretrained_Model_G: 4x_Compact_Pretrain
Training time: 24h+

Description: Compact 2x anime upscaler with otf compression and blur. The '2xHFA2kCompact.pth' (4.6 MB) is the original trained model file, the other model files are conversions using chaiNNer. Trained on musl's latest dataset release for Anime SISR, which has been extracted from modern anime films, where the selection criteria was high SNR, no DOF and high frequency information.

Examples: https://imgsli.com/MTcxNjA4

Example1 Example2 Example3 Example4 Example5 Example6 Example7 Example8 Example9 Example10 Example11 Example12 Example13 Example14 Example15 Example16 Example17 Example18 Example19 Example20 Example21 Example22 Example23 Example24 Example25

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Examples
Unable to determine this model's library. Check the docs .

Spaces using Phips/2xHFA2kCompact 2