File size: 3,616 Bytes
efff97a ac0e820 efff97a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 | ---
license: cc-by-nc-sa-4.0
license_link: https://github.com/wyf0912/SinSR/blob/main/LICENSE
library_name: coreml
pipeline_tag: image-to-image
tags:
- coreml
- core-ml
- ios
- macos
- apple
- on-device
- super-resolution
- diffusion
- one-step
- arxiv:2311.14760
---
# SinSR — Core ML
*Single-Step Super-Resolution, 2024*
4× super-resolution via single-step diffusion. 256→1024. Swin Transformer denoiser (FP32).
<p><img src="https://huggingface.co/mlboydaisuke/SinSR-CoreML/resolve/main/media/75b8b5fc68.png" alt="SinSR demo"></p>
Core ML conversion of [wyf0912/SinSR](https://github.com/wyf0912/SinSR) for on-device inference on iPhone, iPad and Mac. Converted with `coremltools`; the packages are stateless, so all sequencing and buffering lives in your Swift code.
| | |
|---|---|
| Task | image to image |
| Upstream | [wyf0912/SinSR](https://github.com/wyf0912/SinSR) |
| Packages | 3 |
| Download size | 517 MB |
| Minimum iOS | 17.0 |
| Peak RAM | ~600 MB |
## Files
| File | Size | Compute units | SHA-256 |
|---|---:|---|---|
| `SinSR_Encoder.mlpackage.zip` | 39 MB | `cpuAndGPU` | `fdec09d17561ec1b…` |
| `SinSR_Denoiser.mlpackage.zip` | 420 MB | `cpuOnly` | `b31374c2d539b2cd…` |
| `SinSR_Decoder.mlpackage.zip` | 58 MB | `cpuAndGPU` | `b8b9a7b52d6b240c…` |
| **Total** | **517 MB** | | |
`compute_units` is not a suggestion -- it is the configuration the conversion was verified against. Moving a package to a different compute unit can silently change the numerics (FP16 attention overflow) or crash on the GPU.
## Download
```bash
hf download mlboydaisuke/coreml-zoo --include "sinsr/*" --local-dir ./sinsr
unzip './sinsr/sinsr/*.zip' -d ./sinsr
```
## Use in Swift
```swift
import CoreML
let config = MLModelConfiguration()
config.computeUnits = .cpuAndGPU // as converted — see the table above
// Unzip the .mlpackage, drop it into your Xcode target and Xcode compiles it
// at build time:
let model = try SinSR_Encoder(configuration: config)
// ...or compile a downloaded .mlpackage at runtime:
let compiled = try await MLModel.compileModel(at: mlpackageURL)
let model = try MLModel(contentsOf: compiled, configuration: config)
```
> This model is split into 3 Core ML packages that are driven in sequence from Swift. Load them one at a time, copy the outputs out of the `MLMultiArray` buffers and release each model before loading the next — two large Core ML models resident at once will OOM on an iPhone.
## Demo
- **Sample app** — [`sample_apps/SinSRDemo`](https://github.com/john-rocky/CoreML-Models/tree/master/sample_apps/SinSRDemo), a standalone SwiftUI project.
- **Models Zoo** — this model is downloadable and runnable inside the [Models Zoo app](https://apps.apple.com/app/id6762083207) on the App Store, no build required.
## Conversion
- Script: [`convert_sinsr.py`](https://github.com/john-rocky/CoreML-Models/blob/master/conversion_scripts/convert_sinsr.py)
- Pitfalls hit during conversion (FP16 overflow, ANE buffer limits, stride handling): [`docs/coreml_conversion_notes.md`](https://github.com/john-rocky/CoreML-Models/blob/master/docs/coreml_conversion_notes.md)
- Model index: [CoreML-Models](https://github.com/john-rocky/CoreML-Models)
## License
The conversion inherits the upstream license: **CC BY-NC-SA 4.0**.
See [https://github.com/wyf0912/SinSR/blob/main/LICENSE](https://github.com/wyf0912/SinSR/blob/main/LICENSE).
> Non-commercial, attribution, share-alike.
## Credits
- Upstream authors: [wyf0912/SinSR](https://github.com/wyf0912/SinSR), 2024
- Core ML conversion: john-rocky (Daisuke Majima)
|