Instructions to use rossheaton/british-birds-vit-base-patch16-224 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use rossheaton/british-birds-vit-base-patch16-224 with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://rossheaton/british-birds-vit-base-patch16-224") - Notebooks
- Google Colab
- Kaggle
Commit ·
222b7bc
1
Parent(s): bab94c2
Update friendly_class_names.csv to include rspb_description column
Browse files- README.md +1 -1
- model/friendly_class_names.csv +0 -0
README.md
CHANGED
|
@@ -21,7 +21,7 @@ The model was fine-tuned from the `vit_base_patch16_224_imagenet21k` backbone us
|
|
| 21 |
|
| 22 |
## Dataset
|
| 23 |
The dataset comprises **224 classes** with exactly **500 image samples per class** (112,000 images total).
|
| 24 |
-
Images were sourced from [eBird](https://ebird.org/) on February 28, 2026, using the [Birdhouse](https://github.com/rossheat/birdhouse) CLI tool, strictly filtering for the Great Britain region (`--region=GB`),
|
| 25 |
|
| 26 |
The data was split:
|
| 27 |
* **Training:** 80% (89,600 images)
|
|
|
|
| 21 |
|
| 22 |
## Dataset
|
| 23 |
The dataset comprises **224 classes** with exactly **500 image samples per class** (112,000 images total).
|
| 24 |
+
Images were sourced from [eBird](https://ebird.org/) on February 28, 2026, using the [Birdhouse](https://github.com/rossheat/birdhouse) CLI tool, strictly filtering for the Great Britain region (`--region=GB`), moderately rated images (`--min-avg-rating=4`), with a reasonable number of reviews (`--min-reviews=2`).
|
| 25 |
|
| 26 |
The data was split:
|
| 27 |
* **Training:** 80% (89,600 images)
|
model/friendly_class_names.csv
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|