Downloads · 30 days
12
12% of all-time downloads
ghotriw/deberta-v3-small-biaffine-dep-pos-feats-en-ewt-gum
deberta-v3-small-biaffine-dep-pos-feats-en-ewt-gum is a token classification model from ghotriw. Use it when you need labels on individual words, such as names. The card lists the license as mit.
Biaffine dependency parser + joint UPOS + morphological features (FEATS) tagger trained on Universal Dependencies English Web Treebank (EWT) and Universal Dependencies English GUM.
Downloads · 30 days
12
12% of all-time downloads
All-time downloads
97
Public
Repo size
4.9 GB
Likes
0
Public
Click a slice to open those files.
.onnx866 MB · 59%
From the Hugging Face model README
Biaffine dependency parser + joint UPOS + morphological features (FEATS) tagger trained on Universal Dependencies English Web Treebank (EWT) and Universal Dependencies English GUM.
Encoder: microsoft/deberta-v3-small (frozen weights not included — loaded from HuggingFace at runtime)
Custom head: biaffine_head.safetensors — word projection, arc/rel/POS/FEATS MLPs and biaffine layers
Labels: 53 DEPREL labels · 19 UPOS tags · 21 FEATS categories
Score convention: s_arc[dep, head], s_rel[dep, head, rel]
| Split | LAS | UPOS | UFEATS | UCM | LCM |
|---|---|---|---|---|---|
| dev | 93.14% | 98.23% | 95.96% | 71.36% | 61.27% |
| test | 93.06% | 98.19% | 96.18% | 70.40% | 61.20% |
model.onnx — fp32 ONNX model (Recommended for CPU inference).
model.fp16.onnx — fp16 ONNX model (For GPU or environments with native fp16 support, ~275 MB).
traced_model.pt — TorchScript model (For tch-rs or PyTorch C++ API).
Inputs: subwords [B, W, 20] int64. Outputs: s_arc [B,W,W], s_rel [B,W,W,R], s_pos [B,W,P], s_feats [B,W,C,Vmax].
# download only inference artifacts
hf download ghotriw/deberta-v3-small-biaffine-dep-pos-feats-en-ewt-gum \
model.fp16.onnx model.onnx traced_model.pt vocabs.json tokenizer.json \
--local-dir ./model
vocabs.json — DEPREL and UPOS vocabularies (str→int dicts).
feats_vocab — morphological categories {category: {value: idx}} (idx 0 = _/absent). s_feats[..., c, :] is an independent softmax per category; non-existent value slots carry -inf, so argmax over the last dim is always valid.
The model expects a word-level subword grid [B, W, fix_len=20] (int64),
where each word is independently tokenised with the encoder's sentencepiece tokeniser
and padded/truncated to 20 subword slots. Position 0 is a synthetic ROOT word
whose only subword is [CLS] (id 1).
config.json contains rel_vocab (str→int) and pos_vocab (str→int).
Index 0 is the <pad> / ROOT slot and should be ignored in evaluation.
For FEATS models, config.json also contains feats_vocab {category: {value: idx}} and feats_sizes (per-category value counts, incl. the _/absent slot at index 0). The 4th output s_feats [B,W,C,Vmax] is decoded per category via argmax over its last dimension.