README: nano4 additions, checkpoint count
Browse files
README.md
CHANGED
|
@@ -16,7 +16,7 @@ fine-grained classification head, across 21 fine-grained datasets. FGIRFT is the
|
|
| 16 |
counterpart to the ERIS Lab's parameter-efficient works (the ILA/SAW and AAA adapters in
|
| 17 |
`ERISLab/FGIR-ViT`, which keep the backbone frozen and train only small adapters).
|
| 18 |
|
| 19 |
-
|
| 20 |
(111) and `ft_448` (131). Everything is described in `manifest.csv` and loadable by name with the
|
| 21 |
standalone `fgir_zoo` library (no training repo needed).
|
| 22 |
|
|
|
|
| 16 |
counterpart to the ERIS Lab's parameter-efficient works (the ILA/SAW and AAA adapters in
|
| 17 |
`ERISLab/FGIR-ViT`, which keep the backbone frozen and train only small adapters).
|
| 18 |
|
| 19 |
+
246 checkpoints, one seed per configuration, last epoch only, grouped by input resolution: `ft_224`
|
| 20 |
(111) and `ft_448` (131). Everything is described in `manifest.csv` and loadable by name with the
|
| 21 |
standalone `fgir_zoo` library (no training repo needed).
|
| 22 |
|