mage0001 commited on
Commit
85cf5cd
·
verified ·
1 Parent(s): e37456b

Document clean release validation

Browse files
Files changed (3) hide show
  1. README.md +3 -1
  2. SHA256SUMS.json +4 -4
  3. VALIDATION.md +17 -3
README.md CHANGED
@@ -123,7 +123,9 @@ another base. For completely offline use, keep `sapiens2_backbone.py` beside the
123
 
124
  See `VALIDATION.md` for the checks performed before publishing this loader.
125
  The verified standalone loader is included here. Training, evaluation, TailGuard,
126
- and dataset tooling are maintained in the accompanying source repository.
 
 
127
 
128
  ## Attribution
129
 
 
123
 
124
  See `VALIDATION.md` for the checks performed before publishing this loader.
125
  The verified standalone loader is included here. Training, evaluation, TailGuard,
126
+ and dataset tooling are maintained in the accompanying source repository. The
127
+ private source release candidate `v0.1.0-rc2` includes the complete clean HInt
128
+ validation replay and end-to-end TailGuard command.
129
 
130
  ## Attribution
131
 
SHA256SUMS.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "README.md": {
3
- "bytes": 5977,
4
- "sha256": "87b71da40640785d90f95122fab2c6a77e31a399990e4af5dd70dc3324b76eb9"
5
  },
6
  "training/S2HAND-Training-WarmStart-5B/manifest.json": {
7
  "bytes": 1092,
@@ -56,8 +56,8 @@
56
  "sha256": "1c77a629ee16972ddb9844378038b20321143a6faf0cac26c2c04dc9b61c5288"
57
  },
58
  "VALIDATION.md": {
59
- "bytes": 1547,
60
- "sha256": "9033908aac29ebc10fae97036a14f356c57c2cbd4529025c9bf7bf72dfd02482"
61
  },
62
  "requirements.txt": {
63
  "bytes": 51,
 
1
  {
2
  "README.md": {
3
+ "bytes": 6112,
4
+ "sha256": "1e215e3e017e631c30fcbc79467d53618a40f021e886fc570c090a45ed068797"
5
  },
6
  "training/S2HAND-Training-WarmStart-5B/manifest.json": {
7
  "bytes": 1092,
 
56
  "sha256": "1c77a629ee16972ddb9844378038b20321143a6faf0cac26c2c04dc9b61c5288"
57
  },
58
  "VALIDATION.md": {
59
+ "bytes": 2336,
60
+ "sha256": "fd538ec44824b292bede76200b5e12754fef056f3e6dd5ad553c427b5795d2a7"
61
  },
62
  "requirements.txt": {
63
  "bytes": 51,
VALIDATION.md CHANGED
@@ -25,6 +25,20 @@ The standalone image pipeline is also covered by unit tests for:
25
  - RGB/ImageNet input tensor shape and finite normalization; and
26
  - 308-channel UDP/DARK decoding back to original-image coordinates.
27
 
28
- These tests verify loading, architecture, routing, preprocessing, and decoding.
29
- A fresh full-dataset paper-metric replay is still tracked separately and is not
30
- implied by these deterministic reconstruction tests.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25
  - RGB/ImageNet input tensor shape and finite normalization; and
26
  - 308-channel UDP/DARK decoding back to original-image coordinates.
27
 
28
+ On 2026-10-30, the standalone Pose-5B trainer additionally passed a real
29
+ one-step L40S/BF16 optimization and independent checkpoint-reload test. The
30
+ released MT-5B route was checked on a real HInt image: all 268 protected
31
+ heatmaps were bitwise equal to the public route, while all 40 adapted channels
32
+ used the learned route.
33
+
34
+ The clean checkout also replayed all 1,175 hands in the HInt NewDays/EPICK
35
+ validation split. The catastrophic-tail rate matched exactly
36
+ (`0.0144680851`). Selection AUC was `0.7656657143`, versus `0.7656003214` in
37
+ the original research record (absolute difference `0.0000653929`), and
38
+ combined NME differed by `0.0000097603`. This is a near-exact numerical replay,
39
+ not a bitwise metric claim.
40
+
41
+ Finally, TailGuard was exercised with the released MT-5B delta, real NewDays
42
+ images, person detections, and WiLoR-YOLO hand detections. Non-triggered rows
43
+ remained bitwise equal to native predictions, while every triggered row used
44
+ the documented coherent 21-joint conditional crop.