---
license: mit
language:
  - el
  - grc
tags:
  - lemmatization
  - greek
  - ancient-greek
  - modern-greek
  - character-level
  - transformer
  - morphology
  - nlp
library_name: pytorch
---

# Dilemma Model Weights

Character-level transformer for Greek lemmatization, used as the neural fallback in the [Dilemma](https://github.com/ciscoriordan/dilemma) lemmatizer.

## Model Details

- **Architecture:** Encoder-decoder transformer (character-level)
- **Parameters:** 4.2M
- **d_model:** 256
- **Attention heads:** 4
- **Layers:** 3 (encoder and decoder)
- **Feed-forward dim:** 512
- **Vocabulary:** 381 characters (Greek polytonic + special tokens)
- **Training data:** 3.4M form-lemma pairs from Wiktionary inflection tables
- **Multi-task heads:** POS tagging (10 tags), nominal inflection (45 labels), verbal inflection (69 labels)

## Files

| File | Size | Description |
|------|------|-------------|
| `model.pt` | ~16 MB | PyTorch checkpoint (weights + config) |
| `encoder.onnx` | ~7 MB | ONNX encoder for lightweight inference |
| `decoder_step.onnx` | ~10 MB | ONNX decoder for lightweight inference |
| `vocab.json` | ~9 KB | Character vocabulary (char2id / id2char mappings) |

## Usage

This model is used automatically by the Dilemma library as a fallback for forms not found in the 12.3M-entry lookup table or resolved by rule-based morphological analysis. Only about 5% of Greek words reach the transformer.

```bash
pip install dilemma
```

```python
from dilemma import Dilemma
d = Dilemma()

# The transformer is invoked automatically when needed
d.lemmatize("ἐποιήσαντο", lang="grc")  # -> ποιέω
```

### ONNX vs PyTorch

For inference, ONNX Runtime (~50 MB install) and PyTorch (~2 GB install) produce identical results. The ONNX files are provided for environments where a lighter dependency is preferred. PyTorch is only needed for training.

## Training

Trained from scratch in minutes on a single GPU using the `train.py` script in the Dilemma repository:

```bash
python train.py
python export_onnx.py  # Export to ONNX format
```

## License

MIT