README: link MoGe2 ckpt at the data repo's pretrained/ (no longer GDrive)

README.md

@@ -37,7 +37,7 @@ PPD (and therefore LPD) support two semantic-prompt backbones — switching is a
 | Backbone | Trainable params | PPD ckpt | Config |
 |---|---|---|---|
 | **DA2** (Depth-Anything-V2 ViT-L) | 16.3 M | [`gangweix/Pixel-Perfect-Depth/ppd.pth`](https://huggingface.co/gangweix/Pixel-Perfect-Depth/resolve/main/ppd.pth) | `code/ppd/configs/lpd_run5d_10k.yaml` |
-| **MoGe2** | 16.3 M | [Google Drive](https://drive.google.com/file/d/1tabmcsbRVDKDfmO4KU1vOjurzN-wp0HV/view?usp=sharing) | `code/ppd/configs/lpd_run5d_moge2.yaml` |
+| **MoGe2** | 16.3 M | [`chenming-wu/LiDAR-Perfect-Depth-Datasets/pretrained/ppd_moge2/`](https://huggingface.co/datasets/chenming-wu/LiDAR-Perfect-Depth-Datasets/tree/main/pretrained/ppd_moge2) | `code/ppd/configs/lpd_run5d_moge2.yaml` |
 
 The official PPD MoGe2 release reports 20-30 % improvement over DA2 on zero-shot benchmarks; the LPD prompt branch is identical for either backbone.