README.md

---
dataset_info:
  features:
  - name: so
    dtype: string
  - name: en
    dtype: string
  - name: sco
    dtype: float64
  splits:
  - name: train
    num_bytes: 582313668
    num_examples: 2875262
  download_size: 396107095
  dataset_size: 582313668
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
task_categories:
- translation
size_categories:
- 1M<n<10M
language:
- so
- en
---

# `quickmt` `so-en` Training Corpus

Contains the following datasets downloaded with [`mtdata`](https://github.com/thammegowda/mtdata) after basic filtering: 

| Corpus | Count |
|---|---|
| OPUS-tatoeba-v20200531-eng-som | 1 |
| OPUS-qed-v2.0a-eng-som | 1751 |
| OPUS-xlent-v1.1-eng-som | 69982 |
| OPUS-ubuntu-v14.10-eng_GB-som | 28 |
| OPUS-xlent-v1-eng-som | 69982 |
| OPUS-bible_uedin-v1-eng-som | 62195 |
| OPUS-ted2020-v1-eng-som | 1980 |
| AllenAi-nllb-1-eng-som | 10229073 |
| OPUS-paracrawl-v8-eng-som | 14879 |
| OPUS-gnome-v1-eng_CA-som | 783 |
| OPUS-ccaligned-v1-eng-som | 364952 |
| OPUS-tatoeba-v20201109-eng-som | 3 |
| OPUS-infopankki-v1-eng-som | 47220 |
| OPUS-paracrawl-v9-eng-som | 14879 |
| ParaCrawl-paracrawl-1_bonus-eng-som | 14879 |
| OPUS-tico_19-v20201028-eng-som | 3071 |
| OPUS-gnome-v1-eng-som | 753 |
| OPUS-ubuntu-v14.10-eng_AU-som | 28 |
| Statmt-ccaligned-1-eng-som_SO | 364959 |
| OPUS-paracrawl-v7.1-eng-som | 14879 |
| OPUS-wikimedia-v20190628-eng-som | 766 |
| OPUS-ccmatrix-v1-eng-som | 222793 |
| OPUS-gnome-v1-eng_GB-som | 787 |
| OPUS-tanzil-v1-eng-som | 93844 |
| OPUS-wikimedia-v20230407-eng-som | 1126 |
| OPUS-xlent-v1.2-eng-som | 69982 |
| OPUS-tatoeba-v20210722-eng-som | 4 |
| OPUS-tatoeba-v20220303-eng-som | 4 |
| OPUS-wikimedia-v20210402-eng-som | 1013 |
| OPUS-ubuntu-v14.10-eng-som | 28 |
| OPUS-tatoeba-v20210310-eng-som | 4 |
| OPUS-multiccaligned-v1-eng-som | 364959 |
| OPUS-multiparacrawl-v7.1-eng-som | 14879 |