Instructions to use ERISLab/TGDA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- timm
How to use ERISLab/TGDA with timm:
import timm model = timm.create_model("hf-hub:ERISLab/TGDA", pretrained=True) - Notebooks
- Google Colab
- Kaggle
Card: link the ERISLab collection
Browse files
README.md
CHANGED
|
@@ -15,7 +15,8 @@ These are the checkpoints behind *Fine-Grained Image Recognition from Scratch wi
|
|
| 15 |
249 checkpoints, one per configuration, each the last epoch of one training seed. Each file
|
| 16 |
is a `torch.save` dict with `config` (the full training configuration), `model` (the state dict),
|
| 17 |
`accuracy` and `epoch`, with no optimizer state. File names are the runs' experiment-log names,
|
| 18 |
-
ending in the serial. Load them with [fgir-zoo](https://github.com/arkel23/fgir-zoo).
|
|
|
|
| 19 |
|
| 20 |
## Layout
|
| 21 |
|
|
|
|
| 15 |
249 checkpoints, one per configuration, each the last epoch of one training seed. Each file
|
| 16 |
is a `torch.save` dict with `config` (the full training configuration), `model` (the state dict),
|
| 17 |
`accuracy` and `epoch`, with no optimizer state. File names are the runs' experiment-log names,
|
| 18 |
+
ending in the serial. Load them with [fgir-zoo](https://github.com/arkel23/fgir-zoo). The
|
| 19 |
+
[collection](https://huggingface.co/collections/ERISLab/tgda-fgir-from-scratch-with-teacher-guided-augmentation-6ab32cd151f4fc1063b55f64) groups this repo with the paper.
|
| 20 |
|
| 21 |
## Layout
|
| 22 |
|