--- title: "Own models & training" url: https://phatdev.net/en/docs/pixgamma/model-rieng lang: en updated: 2026-10-22 alternate: vi: https://phatdev.net/docs/pixgamma/model-rieng markdown: https://phatdev.net/en/docs/pixgamma/model-rieng.md --- # Own models & training Part of the doc set: [PixGamma](https://phatdev.net/en/docs/pixgamma.md) > PixGamma Photo v2 and v1, the benchmark of public models, versions tried and the training toolkit. PixGamma has its own models, trained on the Mac with [traiNNer-redux](https://github.com/the-database/traiNNer-redux) on the GPU (MPS) and converted to Core ML. A model only ships when it beats the current one on **both benchmarks**: the scored photo set and a full-pipeline run on real photos.
DISTS on 40 photos never used for training: PixGamma Photo v1 against 9 public modelsDISTS on 40 photos never used for training: PixGamma Photo v1 against 9 public models
DISTS on 40 photos never used for training: PixGamma Photo v1 against 9 public models
## Choosing a base model PixGamma scored highly rated public 4x models (Hugging Face, OpenModelDB, GitHub) on 40 photos never used for training: 22 laptop shots plus portraits, animals, fabric and leather. The main metric is **DISTS** (texture similarity, lower is better), alongside PSNR. | 4x model | DISTS | |---|---| | **PixGamma Photo v2** | **0.164** | | PixGamma Photo v1 | 0.165 | | 4xBHI_dat2_real (base of Photo v1) | 0.173 | | UltraSharpV2 Lite | 0.176 | | RealWebPhoto ATD | 0.176 | | UltraSharpV2 | 0.177 | | RealWebPhoto DAT2 | 0.179 | | 4xBHI RealPLKSR | 0.180 | | Real-ESRGAN x4plus | 0.186 | | 4xNomosWebPhoto (Photo model in 0.1) | 0.199 | PSNR on the same set: PixGamma Photo v1 30.5 dB, 4xBHI_dat2_real 29.0 dB, 4xNomosWebPhoto 26.9 dB. ## PixGamma Photo v2 (default photo model)
PixGamma Photo v2: a cat from 480 px to 4x next to the 1920 px original; clearer fur than the input, but not yet strand-for-strand like the originalPixGamma Photo v2: a cat from 480 px to 4x next to the 1920 px original; clearer fur than the input, but not yet strand-for-strand like the original
PixGamma Photo v2: a cat from 480 px to 4x next to the 1920 px original; clearer fur than the input, but not yet strand-for-strand like the original
- Photo v1 trained further with a **light GAN** for crisper fur, hair and fabric grain. - DISTS on the 40-photo set: **0.164** (v1: 0.165); on sharp photos: **0.170** (v1: 0.183). - **Text regions keep Photo v1** (no GAN), so the GAN never draws letter strokes. - A second round with a stronger GAN scored worse and is not shipped. - Photo v2 is a 4x model; 2x upscales still use SPAN. ## PixGamma Photo v1 - **DAT2** architecture, fine-tuned for 15,000 iterations from 4xBHI_dat2_real (Philip Hofmann, CC-BY-4.0). - Data: CC0 and public-domain photos from Wikimedia Commons, with artwork and sensitive images filtered out and the benchmark photos excluded. Half of the low-resolution inputs go through a **downscale + double JPEG chain** like web images; 2,500 synthetic text tiles keep strokes intact. - Losses: Charbonnier and MS-SSIM (pixel accuracy), LDL (artifact suppression), focal frequency (high frequencies), DISTS (real texture). **No GAN**, so the model does not invent texture. - Result: the base DAT2 brightens images by about one gray level and shifts dark areas; v1 shows almost no color drift. ## Tried and not shipped | Version | Tried | Why not | |---|---|---| | First Photo v2 (0.2.1) | v1 plus a light GAN, used on text too | Broken small text, jagged key edges; the current build keeps text regions on Photo v1 | | Photo v2 round 2 | A stronger GAN | Scored worse than the shipped v2 | | Text v4 | Text model with thin white strokes | Sharper but read 4.3 px capitals worse | | Photo v1 for 2x | 4x then shrink to 2x | Lost to SPAN 2x on 5 of 6 photos in the full pipeline | ## Text model 0.2.0 used a dedicated text model (RealPLKSR fine-tuned from 4xNomosWebPhoto in three rounds on synthetic Vietnamese and English text: 114 fonts with Vietnamese, random syllables covering every diacritic combination, banners, labels, keyboards). Since 0.2.1 text regions use Photo v1, which reads better; the text model stays in the app as a fallback. ## Tooling | Script | Does | |---|---| | `training/fetch_commons.py` | Downloads CC0 and public-domain photos from Wikimedia Commons with credits | | `training/prepare_dataset.py` | Cuts tiles and builds low-resolution inputs with a web-like degradation chain | | `training/make_text_dataset.py` | Generates Vietnamese and English text tiles | | `training/evaluate.py` | Scores PSNR, DISTS and color drift | | `training/eval_text.py`, `eval_text_pipeline.py` | Score text with per-line, per-size OCR | | `tools/convert_extra.py` | Converts checkpoints to Core ML and checks the difference against PyTorch | Every base model has an open license (CC-BY-4.0, BSD-3-Clause, Apache-2.0, MIT, OpenRAIL-M); CodeFormer is non-commercial only.