geo_prior v1: 1 epoch imagenet-synthetic-schnell-10k, shift=2.5

README.md

@@ -1,3 +1,72 @@
----
-license: apache-2.0
----
+---
+license: mit
+library_name: sd15-flow-trainer
+tags:
+  - geometric-deep-learning
+  - stable-diffusion
+  - ksimplex
+  - pentachoron
+  - flow-matching
+  - cross-attention-prior
+base_model: sd-legacy/stable-diffusion-v1-5
+pipeline_tag: text-to-image
+---
+
+# KSimplex Geometric Attention Prior
+
+Geometric cross-attention prior for SD1.5 using pentachoron (4-simplex) structures.
+
+## Architecture
+
+| Component | Params |
+|-----------|--------|
+| SD1.5 UNet (frozen) | 859,520,964 |
+| **Geo prior (trained)** | **4,845,725** |
+
+The geometric prior modulates CLIP encoder hidden states through
+4-layer stacked k-simplex attention before they reach
+the 16 cross-attention blocks in the UNet.
+
+## Simplex Configuration
+
+| Parameter | Value |
+|-----------|-------|
+| k (simplex dim) | 4 |
+| Embedding dim | 32 |
+| Feature dim | 768 |
+| Stacked layers | 4 |
+| Attention heads | 8 |
+| Base deformation | 0.25 |
+| Residual blend | learnable |
+| Timestep conditioned | True |
+
+## Usage
+
+```python
+from sd15_trainer_geo.pipeline import load_pipeline, load_geo_from_hub
+
+# Load base SD1.5 + fresh geo prior
+pipe = load_pipeline()
+
+# Load trained geo weights from this repo
+load_geo_from_hub(pipe, "AbstractPhil/sd15-rectified-geometric-matching")
+
+# Or one-shot: load base + geo in one call
+pipe = load_pipeline(geo_repo_id="AbstractPhil/sd15-rectified-geometric-matching")
+```
+
+## Training Info
+
+- **dataset**: AbstractPhil/imagenet-synthetic (flux_schnell_512)
+- **samples**: 10000
+- **epochs**: 1
+- **shift**: 2.5
+- **base_lr**: 0.0001
+- **min_snr_gamma**: 5.0
+- **cfg_dropout**: 0.1
+- **batch_size**: 6
+- **loss_final**: 0.3784324672818184
+
+## License
+
+MIT — [AbstractPhil](https://huggingface.co/AbstractPhil)

geo_prior.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:103aa7f7b25704f22736673cc706573cd97f4e0fef48063560149d278c361f79
+size 19391076

simplex_config.json

@@ -0,0 +1,16 @@
+{
+  "k": 4,
+  "edim": 32,
+  "feat_dim": 768,
+  "num_layers": 4,
+  "base_deformation": 0.25,
+  "learnable_deformation": true,
+  "timestep_conditioned": true,
+  "num_heads": 8,
+  "dropout": 0.0,
+  "cm_loss_weight": 0.01,
+  "vol_consistency_weight": 0.005,
+  "residual_blend": "learnable",
+  "initial_blend": 0.0,
+  "_base_repo": "sd-legacy/stable-diffusion-v1-5"
+}