fix readme.md

README.md

@@ -1,30 +1,83 @@
 ---
+pipeline_tag: fill-mask
 license: mit
-library_name: keras
-pipeline_tag: feature-extraction
+base_model: microsoft/deberta-v3-small
+library_name: kerasformers
 tags:
-  - keras
-  - kerasformers
-  - tf
-  - jax
-  - pytorch
-  - backend-agnostic
-  - deberta_v3
-  - feature-extraction
-base_model:
-  - microsoft/deberta-v3-small
+- keras
+- kerasformers
+- deberta
+- deberta-v3
+- fill-mask
+- text-encoder
+- arxiv:2006.03654
+- arxiv:2111.09543
+- pytorch
+- jax
+- tf
 ---
-# DeBERTa-v3 (deberta_v3_small)
 
-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).
+## ***See [our collection](https://huggingface.co/collections/kerasformers/deberta-v1-v2-v3-6a6e90bac01e412b478562f3) for all versions of DeBERTa (v1 / v2 / v3).***
 
-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).
+# Run DeBERTa with Keras 3: JAX, PyTorch, or TensorFlow
 
-## Usage
+[![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-DeBERTa-blue)](https://imvision12.github.io/KerasFormers/deberta/) [![Collection](https://img.shields.io/badge/HF-DeBERTa%20collection-yellow)](https://huggingface.co/collections/kerasformers/deberta-v1-v2-v3-6a6e90bac01e412b478562f3)
+
+# kerasformers/deberta_v3_small
+
+Papers: [DeBERTa: Decoding-enhanced BERT with Disentangled Attention (arXiv:2006.03654)](https://arxiv.org/abs/2006.03654) · [DeBERTaV3 (arXiv:2111.09543)](https://arxiv.org/abs/2111.09543) · [HF Papers](https://huggingface.co/papers/2006.03654)
+
+DeBERTa is Microsoft's disentangled-attention text encoder (content + relative position). v1 uses byte-level BPE; v2/v3 use SentencePiece. v3 adds ELECTRA-style pretraining with gradient-disentangled embedding sharing. Import from `deberta` / `deberta_v2` / `deberta_v3` to match the generation.
+
+For more details on the model, please go to the upstream [model card](https://huggingface.co/microsoft/deberta-v3-small).
+
+Pure-**Keras 3** conversion of [`microsoft/deberta-v3-small`](https://huggingface.co/microsoft/deberta-v3-small) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
+
+This is a **fill-mask / encoder** checkpoint (`DebertaV3MaskedLM`, v3 small). Task heads (sequence/token classify, QA, …) load via `hf:` fine-tunes.
+
+## ✨ Quick start (fill-mask)
 
 ```python
-from kerasformers.models.deberta_v3 import DebertaV3Model
+import os
+os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"
+
+from kerasformers.models.deberta_v3 import (
+    DebertaV3MaskedLM,
+    DebertaV3Tokenizer,
+)
 
-model = DebertaV3Model.from_weights("kerasformers/deberta_v3_small")
+mlm = DebertaV3MaskedLM.from_weights("kerasformers/deberta_v3_small")
+tokenizer = DebertaV3Tokenizer.from_weights("kerasformers/deberta_v3_small")
+
+inputs = tokenizer("The capital of France is [MASK].")
+logits = mlm(inputs)  # (1, L, vocab_size)
+mask = int((inputs["input_ids"][0] == tokenizer.mask_token_id).argmax())
+print(tokenizer.decode([int(logits[0, mask].argmax())]))
 ```
 
+Load any DeBERTa variant the same way with `from_weights("kerasformers/<variant>")`:
+
+| Variant | Hub | Generation |
+|---|---|---|
+| `deberta_base` | [`kerasformers/deberta_base`](https://huggingface.co/kerasformers/deberta_base) | v1 |
+| `deberta_large` | [`kerasformers/deberta_large`](https://huggingface.co/kerasformers/deberta_large) | v1 |
+| `deberta_v2_xlarge` | [`kerasformers/deberta_v2_xlarge`](https://huggingface.co/kerasformers/deberta_v2_xlarge) | v2 |
+| `deberta_v2_xxlarge` | [`kerasformers/deberta_v2_xxlarge`](https://huggingface.co/kerasformers/deberta_v2_xxlarge) | v2 |
+| `deberta_v3_xsmall` | [`kerasformers/deberta_v3_xsmall`](https://huggingface.co/kerasformers/deberta_v3_xsmall) | v3 |
+| `deberta_v3_small` | [`kerasformers/deberta_v3_small`](https://huggingface.co/kerasformers/deberta_v3_small) | v3 |
+| `deberta_v3_base` | [`kerasformers/deberta_v3_base`](https://huggingface.co/kerasformers/deberta_v3_base) | v3 |
+| `deberta_v3_large` | [`kerasformers/deberta_v3_large`](https://huggingface.co/kerasformers/deberta_v3_large) | v3 |
+
+## Tips
+
+- Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
+- Prefer `Tokenizer.from_weights(...)` so vocab and mask token match.
+- Do not mix packages across generations (v1 ≠ v2 ≠ v3).
+- See [DeBERTa docs](https://imvision12.github.io/KerasFormers/deberta/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
+- Community / upstream safetensors still work via the `hf:` prefix, e.g. `DebertaV3MaskedLM.from_weights("hf:microsoft/deberta-v3-small")`.
+
+## Special Thanks
+
+A huge thank you to the Microsoft DeBERTa authors for creating and releasing these models.
+
+License: MIT.