CPGA-Net low-light LiteRT fp16 (fully-GPU, Pixel 8a corr 0.99999, 2ms, 0.1MB)

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+samples/sample.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,47 @@
+---
+license: mit
+library_name: LiteRT
+pipeline_tag: image-to-image
+tags: [litert, tflite, on-device, android, gpu, low-light, image-enhancement, cpga-net]
+base_model: Shyandram/CPGA-Net-Pytorch
+---
+
+# CPGA-Net — LiteRT (on-device low-light enhancement, fully-GPU)
+
+[CPGA-Net](https://github.com/Shyandram/CPGA-Net-Pytorch) (Shyandram, IJPRAI), low-light image enhancement
+(brighten dark photos) via Channel Prior + Gamma Correction, converted to **LiteRT** and running **fully on the
+`CompiledModel` GPU** (ML Drift) on Android. **0.025 M params / 0.1 MB fp16 — the smallest model in the
+[LiteRT-Models zoo](https://github.com/john-rocky/LiteRT-Models).**
+
+![CPGA-Net — dark -> enhanced on-device LiteRT GPU](samples/sample.png)
+
+## On-device (Pixel 8a, Tensor G3 — verified)
+
+| | |
+|---|---|
+| nodes on GPU | **135 / 135** LITERT_CL (full residency) |
+| inference | **~2 ms** (256×256) |
+| size | **0.1 MB** (fp16) |
+| accuracy | device-vs-PyTorch corr **0.99999**, no NaN |
+
+```
+image[1,3,256,256] (RGB [0,1]) →[GPU: CPGA-Net]→ enhanced[1,3,256,256] ([0,1])
+```
+
+## How it converts (litert-torch) — three numerically-exact fixes
+
+1. **Gamma correction `x^γ` → `exp(γ·log x)`** — `torch.pow` lowers to the banned `POW`; the identity
+   `x^γ = exp(γ·log x)` is exact (base clamped to [1e-9,1]) and uses native `EXP`/`LOG`.
+2. **CBAM / gamma global pools** → `mean(3).mean(2)` and `F.max_pool2d(x, (H,W))`.
+3. The dark/bright **channel prior** (`max`/`min` over RGB) stays as `REDUCE_MAX`/`REDUCE_MIN`.
+
+Banned ops NONE, ≤4D, tflite-vs-torch corr **1.0**, device-vs-torch corr **0.99999**. Guided-filter disabled.
+
+## Preprocessing
+
+Center-crop, resize 256×256, RGB scaled to [0,1], NCHW.
+
+## License
+
+[MIT](https://github.com/Shyandram/CPGA-Net-Pytorch/blob/main/LICENSE). Upstream:
+[Shyandram/CPGA-Net-Pytorch](https://github.com/Shyandram/CPGA-Net-Pytorch).

cpga_fp16.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b125569618cbc342b3c0a095f712dfc899ac739e896208baf13cb1769c4c319
+size 93820