mlboydaisuke commited on
Commit
7087fdb
·
verified ·
1 Parent(s): 0449f30

Link the card back to its collection and the request box

Browse files
Files changed (1) hide show
  1. README.md +10 -0
README.md CHANGED
@@ -57,3 +57,13 @@ torch.export -> to_edge_transform_and_lower(partitioner) -> .pte
57
  (conversion scripts: [executorch-models](https://github.com/john-rocky/executorch-models))
58
 
59
  **Notes (int8)**: int8 is measured but not shipped. Correlation is a weak read on a six-element output, so the check that matters is the angle between the fp32 and int8 rotations: median 46.5 deg over ten faces, worst 104 deg. That is the known failure mode of re-parameterized RepVGG under post-training quantization — its fused branches leave weight ranges too wide for an int8 grid — and it needs quantization-aware training rather than anything on the export side.
 
 
 
 
 
 
 
 
 
 
 
57
  (conversion scripts: [executorch-models](https://github.com/john-rocky/executorch-models))
58
 
59
  **Notes (int8)**: int8 is measured but not shipped. Correlation is a weak read on a six-element output, so the check that matters is the angle between the fp32 and int8 rotations: median 46.5 deg over ten faces, worst 104 deg. That is the known failure mode of re-parameterized RepVGG under post-training quantization — its fused branches leave weight ranges too wide for an int8 grid — and it needs quantization-aware training rather than anything on the export side.
60
+
61
+ <!-- funnel:v1 -->
62
+
63
+ ---
64
+
65
+ **More models in this format:** [ExecuTorch Model Zoo](https://huggingface.co/collections/mlboydaisuke/executorch-model-zoo-6a7ff328390b63075ffeae5e) — 31 models, each with the recipe that produced it.
66
+
67
+ **Want a different model on-device?** [Open a request](https://github.com/john-rocky/on-device-requests) — free, open weights only; the export and its measured numbers get published publicly.
68
+
69
+ <!-- /funnel:v1 -->