Self-trained models

[Phhofm]

Through your VCISR repo I had also found this inference repo/code, nice :)

I simply wanted to train two models for it and show them here, so I made a Real-CUGAN model, and a Shallow ESRGAN model. But these have been trained with the Real-ESRGAN otf pipeline only and therefore cannot handle video compression / have not been trained for video compression in their current state.

---

Name: 2xHFA2Real-CUGAN
Download Folder
License: CC BY 4.0
Network: Real-CUGAN
Scale: 2
Purpose: 2x anime upscaler
Iterations: 151'000
epoch: 69
batch_size: 6
HR_size: 128 - 256
Dataset: hfa2k
Number of train images: 2568
OTF Training: Yes
Pretrained_Model_G: up2x-latest-conservative

Slow Pics examples:
Example 1
Example 2
Ludvae1
Ludvae2

---

Name: 2xHFA2kShallowESRGAN
Download Folder
License: CC BY 4.0
Network: Shallow ESRGAN (6 Blocks)
Scale: 2
Purpose: 2x anime upscaler
Iterations: 180'000
epoch: 167
batch_size: 12
HR_size: 128
Dataset: hfa2k
Number of train images: 2568
OTF Training: Yes
Pretrained_Model_G: None

Slow Pics examples:
Example 1
Example 2
Ludvae1
Ludvae2

---