The following ready-to-use models are available. For each language, treebanks were shuffled and split 7/1/2 in train/development/held-out partitions. Reported accuracies are from evaluation on held-out data.
Performance in sentences per second is measured using:
- CPU: Ryzen 3700X CPU, using 4 threads.
- GPU: NVIDIA RTX 2060
Models can be used by unpacking the model archive and running
$ syntaxdot annotate model-name/syntaxdot.conf| Model | UD POS | Lemma | UD morph | LAS | Size (MiB) | CPU sent/s | GPU sents/s |
|---|---|---|---|---|---|---|---|
| Finetuned XLM-R base | 98.90 | 99.03 | 98.87 | 94.37 | 1087 | 61 | 755 |
| Distilled, 12 layers, 384 hidden | 98.83 | 99.03 | 98.80 | 93.91 | 200 | 135 | 1450 |
| Distilled, 6 layers, 384 hidden | 98.80 | 99.05 | 98.79 | 93.42 | 133 | 240 | 2359 |
| Model | UD POS | STTS POS | Lemma | UD morph | TDZ morph | LAS | Topo Field | Size (MiB) | CPU sent/s | GPU sent/s |
|---|---|---|---|---|---|---|---|---|---|---|
| Finetuned XLM-R base | 99.54 | 99.48 | 99.34 | 98.38 | 98.43 | 96.59 | 98.17 | 1087 | 45 | 614 |
| Distilled, 12 layers, 384 hidden | 99.50 | 99.44 | 99.31 | 98.31 | 98.36 | 96.17 | 98.12 | 208 | 105 | 1131 |
| Distilled, 6 layers, 384 hidden | 99.46 | 99.40 | 99.29 | 98.20 | 98.27 | 95.48 | 97.97 | 140 | 180 | 1748 |