Update README.md
Browse files
README.md
CHANGED
|
@@ -37,12 +37,11 @@ result.forces # (T, 3) in bodyweight (BW) units; also .time, .sigma
|
|
| 37 |
`trial.npz` is a preprocessed SMPL sequence; see the repo for the video → features pipeline.
|
| 38 |
|
| 39 |
## Intended use & limitations
|
| 40 |
-
Research use only. Trained on the OrthoLoad
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
outside these tested conditions (other activities or populations)
|
| 45 |
-
is not guaranteed.
|
| 46 |
|
| 47 |
## Citation
|
| 48 |
```bibtex
|
|
|
|
| 37 |
`trial.npz` is a preprocessed SMPL sequence; see the repo for the video → features pipeline.
|
| 38 |
|
| 39 |
## Intended use & limitations
|
| 40 |
+
Research use only. Trained on the OrthoLoad instrumented-implant cohort (26 patients, 25 activities) and evaluated zero-shot
|
| 41 |
+
on an independent cohort (the Grand Challenge knee load competition data) where it matches or outperforms prior published methods.
|
| 42 |
+
This is the **all-subjects** checkpoint (the one used for the paper's inverse design experiments);
|
| 43 |
+
the paper's reported accuracy is **leave-one-subject-out**, so in-sample subjects look better than those held-out figures.
|
| 44 |
+
Accuracy outside these tested conditions (other activities or populations) is not guaranteed.
|
|
|
|
| 45 |
|
| 46 |
## Citation
|
| 47 |
```bibtex
|