|
--- |
|
tags: |
|
- spacy |
|
- floret |
|
- fasttext |
|
- feature-extraction |
|
- token-classification |
|
language: |
|
- hu |
|
license: cc-by-sa-4.0 |
|
model-index: |
|
- name: hu_vectors_web_lg |
|
results: |
|
- task: |
|
name: Analogical questions |
|
type: token-classification |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.1094 |
|
- name: MRR |
|
type: mrr |
|
value: 0.2107 |
|
|
|
--- |
|
Hungarian word vectors for HuSpaCy. |
|
|
|
The model is trained on the Hungarian Webcorpus 2.0 using floret with the following hyperparameters: `floret cbow -dim 300 -mode floret -bucket 200000 -minn 4 -maxn 6 -minCount 100 -neg 10 -hashCount 2 -lr 0.01 -thread 70 -epoch 40` |
|
|
|
Vectors are published in fasttext and floret format. |
|
|
|
| Feature | Description | |
|
| --- | --- | |
|
| **Name** | `hu_vectors_web_lg` | |
|
| **Version** | `1.0` | |
|
| **Vectors** | 200000 keys (300 dimensions) | |
|
| **Sources** | [Hungarian Webcorpus 2.0](https://hlt.bme.hu/en/resources/webcorpus2) (Dávid Márk Nemeskey (SZTAKI-HLT)) | |
|
| **License** | `cc-by-sa-4.0` | |
|
| **Author** | [SzegedAI, MILAB](https://github.com/huspacy/huspacy) | |
|
|
|
|
|
### Accuracy |
|
|
|
| Type | Score | |
|
| --- | --- | |
|
| `ACC` | 10.94 | |
|
| `MRR` | 0.2107 | |
