Upload folder using huggingface_hub
Browse files- README.md +30 -0
- model.weights.h5 +3 -0
- tokenizer.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: keras
|
| 4 |
+
pipeline_tag: feature-extraction
|
| 5 |
+
tags:
|
| 6 |
+
- keras
|
| 7 |
+
- kerasformers
|
| 8 |
+
- tf
|
| 9 |
+
- jax
|
| 10 |
+
- pytorch
|
| 11 |
+
- backend-agnostic
|
| 12 |
+
- deberta_v3
|
| 13 |
+
- feature-extraction
|
| 14 |
+
base_model:
|
| 15 |
+
- microsoft/deberta-v3-small
|
| 16 |
+
---
|
| 17 |
+
# DeBERTa-v3 (deberta_v3_small)
|
| 18 |
+
|
| 19 |
+
A pure [Keras 3](https://keras.io) port of **DeBERTa-v3**, converted from [`microsoft/deberta-v3-small`](https://huggingface.co/microsoft/deberta-v3-small) (Hugging Face Transformers).
|
| 20 |
+
|
| 21 |
+
The weights are **backend-agnostic**: the same checkpoint loads and runs identically under the JAX, TensorFlow, or PyTorch Keras backend (`KERAS_BACKEND=jax|tensorflow|torch`), via [kerasformers](https://github.com/IMvision12/KerasFormers).
|
| 22 |
+
|
| 23 |
+
## Usage
|
| 24 |
+
|
| 25 |
+
```python
|
| 26 |
+
from kerasformers.models.deberta_v3 import DebertaV3Model
|
| 27 |
+
|
| 28 |
+
model = DebertaV3Model.from_weights("kerasformers/deberta_v3_small")
|
| 29 |
+
```
|
| 30 |
+
|
model.weights.h5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:111e394a4753b0ca74e7acc4e376c1f21038bf62e199cbb989c9da4b0888861a
|
| 3 |
+
size 565460752
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|