---
license: cc-by-nc-4.0
task_categories:
- translation
language:
- am
- ar
- ay
- bm
- bbj
- bn
- bg
- ca
- cs
- ku
- da
- de
- el
- en
- et
- ee
- fil
- fi
- fr
- fon
- gu
- ha
- he
- hi
- hu
- ig
- id
- it
- ja
- kk
- km
- ko
- lv
- lt
- lg
- luo
- mk
- mos
- my
- nl
- ne
- or
- pa
- pcm
- fa
- pl
- pt
- mg
- ro
- ru
- es
- sr
- sq
- sw
- sv
- tet
- tn
- tr
- tw
- ur
- wo
- yo
- zh
- zu
multilinguality:
- translation
- multilingual
pretty_name: PolyNewsParallel
size_categories:
- 1K<n<10K
source_datasets:
- mafand
- wmt-news
- globalvoices
tags:
- news
- polynews-parallel
- mafand
- globalvoices
- wmtnews
configs:
- config_name: ces_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-tur_Latn/train.parquet.gzip"
- config_name: mya_Mymr-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/mya_Mymr-rus_Cyrl/train.parquet.gzip"
- config_name: plt_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-nld_Latn/train.parquet.gzip"
- config_name: hun_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/hun_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: bul_Cyrl-swh_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-swh_Latn/train.parquet.gzip"
- config_name: amh_Ethi-deu_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-deu_Latn/train.parquet.gzip"
- config_name: cat_Latn-ell_Grek
  data_files: 
  - split: train 
    path: "data/cat_Latn-ell_Grek/train.parquet.gzip"
- config_name: cat_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-nld_Latn/train.parquet.gzip"
- config_name: deu_Latn-eng_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-eng_Latn/train.parquet.gzip"
- config_name: ben_Beng-tet_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-tet_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-srp_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-srp_Latn/train.parquet.gzip"
- config_name: arb_Arab-tur_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-tur_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-ita_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-ita_Latn/train.parquet.gzip"
- config_name: ayr_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-plt_Latn/train.parquet.gzip"
- config_name: hin_Deva-ita_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-ita_Latn/train.parquet.gzip"
- config_name: cat_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-hun_Latn/train.parquet.gzip"
- config_name: cat_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/cat_Latn-npi_Deva/train.parquet.gzip"
- config_name: ces_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-ind_Latn/train.parquet.gzip"
- config_name: ces_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-nld_Latn/train.parquet.gzip"
- config_name: arb_Arab-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/arb_Arab-jpn_Jpan/train.parquet.gzip"
- config_name: eng_Latn-ibo_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-ibo_Latn/train.parquet.gzip"
- config_name: ben_Beng-cat_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-cat_Latn/train.parquet.gzip"
- config_name: srp_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/srp_Latn-tur_Latn/train.parquet.gzip"
- config_name: ben_Beng-swh_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-swh_Latn/train.parquet.gzip"
- config_name: deu_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-ron_Latn/train.parquet.gzip"
- config_name: heb_Hebr-ita_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-ita_Latn/train.parquet.gzip"
- config_name: pes_Arab-srp_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-srp_Latn/train.parquet.gzip"
- config_name: eng_Latn-fin_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-fin_Latn/train.parquet.gzip"
- config_name: ben_Beng-heb_Hebr
  data_files: 
  - split: train 
    path: "data/ben_Beng-heb_Hebr/train.parquet.gzip"
- config_name: bul_Cyrl-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-jpn_Jpan/train.parquet.gzip"
- config_name: kor_Hang-zho_Hans
  data_files: 
  - split: train 
    path: "data/kor_Hang-zho_Hans/train.parquet.gzip"
- config_name: nld_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/nld_Latn-zho_Hant/train.parquet.gzip"
- config_name: hun_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-ron_Latn/train.parquet.gzip"
- config_name: npi_Deva-pol_Latn
  data_files: 
  - split: train 
    path: "data/npi_Deva-pol_Latn/train.parquet.gzip"
- config_name: ayr_Latn-bul_Cyrl
  data_files: 
  - split: train 
    path: "data/ayr_Latn-bul_Cyrl/train.parquet.gzip"
- config_name: ita_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/ita_Latn-urd_Arab/train.parquet.gzip"
- config_name: ayr_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/ayr_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: ces_Latn-heb_Hebr
  data_files: 
  - split: train 
    path: "data/ces_Latn-heb_Hebr/train.parquet.gzip"
- config_name: ayr_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-ron_Latn/train.parquet.gzip"
- config_name: mya_Mymr-sqi_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-sqi_Latn/train.parquet.gzip"
- config_name: fil_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/fil_Latn-urd_Arab/train.parquet.gzip"
- config_name: sqi_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/sqi_Latn-srp_Latn/train.parquet.gzip"
- config_name: por_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-tur_Latn/train.parquet.gzip"
- config_name: plt_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-por_Latn/train.parquet.gzip"
- config_name: ben_Beng-tur_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-tur_Latn/train.parquet.gzip"
- config_name: khm_Khmr-zho_Hant
  data_files: 
  - split: train 
    path: "data/khm_Khmr-zho_Hant/train.parquet.gzip"
- config_name: ory_Orya-urd_Arab
  data_files: 
  - split: train 
    path: "data/ory_Orya-urd_Arab/train.parquet.gzip"
- config_name: ben_Beng-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/ben_Beng-mkd_Cyrl/train.parquet.gzip"
- config_name: eng_Latn-lug_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-lug_Latn/train.parquet.gzip"
- config_name: hun_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-swh_Latn/train.parquet.gzip"
- config_name: spa_Latn-ckb_Arab
  data_files: 
  - split: train 
    path: "data/spa_Latn-ckb_Arab/train.parquet.gzip"
- config_name: por_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-srp_Latn/train.parquet.gzip"
- config_name: kor_Hang-nld_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-nld_Latn/train.parquet.gzip"
- config_name: amh_Ethi-zho_Hans
  data_files: 
  - split: train 
    path: "data/amh_Ethi-zho_Hans/train.parquet.gzip"
- config_name: ron_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/ron_Latn-swe_Latn/train.parquet.gzip"
- config_name: dan_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/dan_Latn-kor_Hang/train.parquet.gzip"
- config_name: amh_Ethi-nld_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-nld_Latn/train.parquet.gzip"
- config_name: ita_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ita_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: jpn_Jpan-ory_Orya
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-ory_Orya/train.parquet.gzip"
- config_name: ayr_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-ita_Latn/train.parquet.gzip"
- config_name: eng_Latn-pcm_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-pcm_Latn/train.parquet.gzip"
- config_name: ben_Beng-khm_Khmr
  data_files: 
  - split: train 
    path: "data/ben_Beng-khm_Khmr/train.parquet.gzip"
- config_name: ita_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/ita_Latn-ory_Orya/train.parquet.gzip"
- config_name: hin_Deva-mya_Mymr
  data_files: 
  - split: train 
    path: "data/hin_Deva-mya_Mymr/train.parquet.gzip"
- config_name: deu_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/deu_Latn-khm_Khmr/train.parquet.gzip"
- config_name: nld_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-swe_Latn/train.parquet.gzip"
- config_name: spa_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-sqi_Latn/train.parquet.gzip"
- config_name: ita_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-swe_Latn/train.parquet.gzip"
- config_name: fil_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/fil_Latn-zho_Hans/train.parquet.gzip"
- config_name: plt_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-yor_Latn/train.parquet.gzip"
- config_name: ind_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-pol_Latn/train.parquet.gzip"
- config_name: amh_Ethi-ell_Grek
  data_files: 
  - split: train 
    path: "data/amh_Ethi-ell_Grek/train.parquet.gzip"
- config_name: fil_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-por_Latn/train.parquet.gzip"
- config_name: ces_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/ces_Latn-zho_Hant/train.parquet.gzip"
- config_name: npi_Deva-swh_Latn
  data_files: 
  - split: train 
    path: "data/npi_Deva-swh_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-nld_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-nld_Latn/train.parquet.gzip"
- config_name: deu_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/deu_Latn-urd_Arab/train.parquet.gzip"
- config_name: srp_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/srp_Latn-swe_Latn/train.parquet.gzip"
- config_name: khm_Khmr-zho_Hans
  data_files: 
  - split: train 
    path: "data/khm_Khmr-zho_Hans/train.parquet.gzip"
- config_name: plt_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-swe_Latn/train.parquet.gzip"
- config_name: ind_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/ind_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: dan_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-nld_Latn/train.parquet.gzip"
- config_name: heb_Hebr-zho_Hant
  data_files: 
  - split: train 
    path: "data/heb_Hebr-zho_Hant/train.parquet.gzip"
- config_name: ell_Grek-tur_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-tur_Latn/train.parquet.gzip"
- config_name: cat_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/cat_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: bul_Cyrl-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-rus_Cyrl/train.parquet.gzip"
- config_name: deu_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-plt_Latn/train.parquet.gzip"
- config_name: rus_Cyrl-swh_Latn
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-swh_Latn/train.parquet.gzip"
- config_name: rus_Cyrl-zho_Hans
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-zho_Hans/train.parquet.gzip"
- config_name: pes_Arab-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/pes_Arab-rus_Cyrl/train.parquet.gzip"
- config_name: ell_Grek-ind_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-ind_Latn/train.parquet.gzip"
- config_name: urd_Arab-zho_Hant
  data_files: 
  - split: train 
    path: "data/urd_Arab-zho_Hant/train.parquet.gzip"
- config_name: ind_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-ita_Latn/train.parquet.gzip"
- config_name: pes_Arab-urd_Arab
  data_files: 
  - split: train 
    path: "data/pes_Arab-urd_Arab/train.parquet.gzip"
- config_name: ayr_Latn-eng_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-eng_Latn/train.parquet.gzip"
- config_name: pol_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/pol_Latn-zho_Hant/train.parquet.gzip"
- config_name: fra_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/fra_Latn-zho_Hant/train.parquet.gzip"
- config_name: ces_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-swe_Latn/train.parquet.gzip"
- config_name: ind_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-swe_Latn/train.parquet.gzip"
- config_name: ayr_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-hun_Latn/train.parquet.gzip"
- config_name: amh_Ethi-hun_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-hun_Latn/train.parquet.gzip"
- config_name: cat_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-tur_Latn/train.parquet.gzip"
- config_name: cat_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-plt_Latn/train.parquet.gzip"
- config_name: dan_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-fra_Latn/train.parquet.gzip"
- config_name: deu_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/deu_Latn-ory_Orya/train.parquet.gzip"
- config_name: fra_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/fra_Latn-kor_Hang/train.parquet.gzip"
- config_name: ben_Beng-ita_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-ita_Latn/train.parquet.gzip"
- config_name: arb_Arab-deu_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-deu_Latn/train.parquet.gzip"
- config_name: por_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-tet_Latn/train.parquet.gzip"
- config_name: plt_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/plt_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: ita_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-tet_Latn/train.parquet.gzip"
- config_name: dan_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-plt_Latn/train.parquet.gzip"
- config_name: arb_Arab-hun_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-hun_Latn/train.parquet.gzip"
- config_name: eng_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/eng_Latn-npi_Deva/train.parquet.gzip"
- config_name: ita_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/ita_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: ory_Orya-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ory_Orya-rus_Cyrl/train.parquet.gzip"
- config_name: fil_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-ind_Latn/train.parquet.gzip"
- config_name: ell_Grek-hun_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-hun_Latn/train.parquet.gzip"
- config_name: deu_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-ita_Latn/train.parquet.gzip"
- config_name: ben_Beng-srp_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-srp_Latn/train.parquet.gzip"
- config_name: hin_Deva-pan_Guru
  data_files: 
  - split: train 
    path: "data/hin_Deva-pan_Guru/train.parquet.gzip"
- config_name: ita_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-ron_Latn/train.parquet.gzip"
- config_name: hin_Deva-zho_Hant
  data_files: 
  - split: train 
    path: "data/hin_Deva-zho_Hant/train.parquet.gzip"
- config_name: hin_Deva-swh_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-swh_Latn/train.parquet.gzip"
- config_name: heb_Hebr-swh_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-swh_Latn/train.parquet.gzip"
- config_name: ces_Latn-fil_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-fil_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-hin_Deva
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-hin_Deva/train.parquet.gzip"
- config_name: ell_Grek-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ell_Grek-rus_Cyrl/train.parquet.gzip"
- config_name: rus_Cyrl-urd_Arab
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-urd_Arab/train.parquet.gzip"
- config_name: nld_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-sqi_Latn/train.parquet.gzip"
- config_name: deu_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-por_Latn/train.parquet.gzip"
- config_name: deu_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-ind_Latn/train.parquet.gzip"
- config_name: ben_Beng-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ben_Beng-rus_Cyrl/train.parquet.gzip"
- config_name: eng_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/eng_Latn-khm_Khmr/train.parquet.gzip"
- config_name: fra_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-swe_Latn/train.parquet.gzip"
- config_name: heb_Hebr-por_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-por_Latn/train.parquet.gzip"
- config_name: swh_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/swh_Latn-tet_Latn/train.parquet.gzip"
- config_name: amh_Ethi-hin_Deva
  data_files: 
  - split: train 
    path: "data/amh_Ethi-hin_Deva/train.parquet.gzip"
- config_name: npi_Deva-nld_Latn
  data_files: 
  - split: train 
    path: "data/npi_Deva-nld_Latn/train.parquet.gzip"
- config_name: ben_Beng-hin_Deva
  data_files: 
  - split: train 
    path: "data/ben_Beng-hin_Deva/train.parquet.gzip"
- config_name: ell_Grek-heb_Hebr
  data_files: 
  - split: train 
    path: "data/ell_Grek-heb_Hebr/train.parquet.gzip"
- config_name: rus_Cyrl-tur_Latn
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-tur_Latn/train.parquet.gzip"
- config_name: ayr_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-tur_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-mya_Mymr
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-mya_Mymr/train.parquet.gzip"
- config_name: eng_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/eng_Latn-zho_Hans/train.parquet.gzip"
- config_name: khm_Khmr-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/khm_Khmr-rus_Cyrl/train.parquet.gzip"
- config_name: ayr_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-tet_Latn/train.parquet.gzip"
- config_name: ell_Grek-swe_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-swe_Latn/train.parquet.gzip"
- config_name: eng_Latn-tsn_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-tsn_Latn/train.parquet.gzip"
- config_name: ces_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-hun_Latn/train.parquet.gzip"
- config_name: ben_Beng-spa_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-spa_Latn/train.parquet.gzip"
- config_name: ces_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-srp_Latn/train.parquet.gzip"
- config_name: fra_Latn-hin_Deva
  data_files: 
  - split: train 
    path: "data/fra_Latn-hin_Deva/train.parquet.gzip"
- config_name: eng_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/eng_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: amh_Ethi-pes_Arab
  data_files: 
  - split: train 
    path: "data/amh_Ethi-pes_Arab/train.parquet.gzip"
- config_name: srp_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/srp_Latn-urd_Arab/train.parquet.gzip"
- config_name: fil_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-hun_Latn/train.parquet.gzip"
- config_name: amh_Ethi-ita_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-ita_Latn/train.parquet.gzip"
- config_name: ckb_Arab-plt_Latn
  data_files: 
  - split: train 
    path: "data/ckb_Arab-plt_Latn/train.parquet.gzip"
- config_name: pes_Arab-sqi_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-sqi_Latn/train.parquet.gzip"
- config_name: pan_Guru-zho_Hans
  data_files: 
  - split: train 
    path: "data/pan_Guru-zho_Hans/train.parquet.gzip"
- config_name: por_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-ron_Latn/train.parquet.gzip"
- config_name: rus_Cyrl-swe_Latn
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-swe_Latn/train.parquet.gzip"
- config_name: dan_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/dan_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: ces_Latn-pes_Arab
  data_files: 
  - split: train 
    path: "data/ces_Latn-pes_Arab/train.parquet.gzip"
- config_name: ben_Beng-urd_Arab
  data_files: 
  - split: train 
    path: "data/ben_Beng-urd_Arab/train.parquet.gzip"
- config_name: kor_Hang-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/kor_Hang-mkd_Cyrl/train.parquet.gzip"
- config_name: swh_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/swh_Latn-tur_Latn/train.parquet.gzip"
- config_name: pan_Guru-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/pan_Guru-rus_Cyrl/train.parquet.gzip"
- config_name: deu_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-srp_Latn/train.parquet.gzip"
- config_name: ita_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-tur_Latn/train.parquet.gzip"
- config_name: pan_Guru-tur_Latn
  data_files: 
  - split: train 
    path: "data/pan_Guru-tur_Latn/train.parquet.gzip"
- config_name: dan_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/dan_Latn-urd_Arab/train.parquet.gzip"
- config_name: arb_Arab-ita_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-ita_Latn/train.parquet.gzip"
- config_name: cat_Latn-fil_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-fil_Latn/train.parquet.gzip"
- config_name: amh_Ethi-ben_Beng
  data_files: 
  - split: train 
    path: "data/amh_Ethi-ben_Beng/train.parquet.gzip"
- config_name: ayr_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-ind_Latn/train.parquet.gzip"
- config_name: arb_Arab-pes_Arab
  data_files: 
  - split: train 
    path: "data/arb_Arab-pes_Arab/train.parquet.gzip"
- config_name: amh_Ethi-cat_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-cat_Latn/train.parquet.gzip"
- config_name: arb_Arab-kor_Hang
  data_files: 
  - split: train 
    path: "data/arb_Arab-kor_Hang/train.parquet.gzip"
- config_name: spa_Latn-heb_Hebr
  data_files: 
  - split: train 
    path: "data/spa_Latn-heb_Hebr/train.parquet.gzip"
- config_name: bul_Cyrl-hun_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-hun_Latn/train.parquet.gzip"
- config_name: srp_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/srp_Latn-zho_Hant/train.parquet.gzip"
- config_name: arb_Arab-ind_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-ind_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-urd_Arab
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-urd_Arab/train.parquet.gzip"
- config_name: ind_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-plt_Latn/train.parquet.gzip"
- config_name: dan_Latn-pes_Arab
  data_files: 
  - split: train 
    path: "data/dan_Latn-pes_Arab/train.parquet.gzip"
- config_name: mkd_Cyrl-urd_Arab
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-urd_Arab/train.parquet.gzip"
- config_name: ayr_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/ayr_Latn-npi_Deva/train.parquet.gzip"
- config_name: jpn_Jpan-por_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-por_Latn/train.parquet.gzip"
- config_name: heb_Hebr-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/heb_Hebr-jpn_Jpan/train.parquet.gzip"
- config_name: kor_Hang-por_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-por_Latn/train.parquet.gzip"
- config_name: cat_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-srp_Latn/train.parquet.gzip"
- config_name: ben_Beng-nld_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-nld_Latn/train.parquet.gzip"
- config_name: ita_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-srp_Latn/train.parquet.gzip"
- config_name: pan_Guru-pol_Latn
  data_files: 
  - split: train 
    path: "data/pan_Guru-pol_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-khm_Khmr
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-khm_Khmr/train.parquet.gzip"
- config_name: amh_Ethi-pol_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-pol_Latn/train.parquet.gzip"
- config_name: ayr_Latn-ell_Grek
  data_files: 
  - split: train 
    path: "data/ayr_Latn-ell_Grek/train.parquet.gzip"
- config_name: arb_Arab-zho_Hant
  data_files: 
  - split: train 
    path: "data/arb_Arab-zho_Hant/train.parquet.gzip"
- config_name: fra_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/fra_Latn-mya_Mymr/train.parquet.gzip"
- config_name: fra_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-tet_Latn/train.parquet.gzip"
- config_name: deu_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/deu_Latn-pan_Guru/train.parquet.gzip"
- config_name: cat_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-swh_Latn/train.parquet.gzip"
- config_name: ayr_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-swh_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-por_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-por_Latn/train.parquet.gzip"
- config_name: eng_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-tur_Latn/train.parquet.gzip"
- config_name: ayr_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/ayr_Latn-zho_Hant/train.parquet.gzip"
- config_name: hun_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-ind_Latn/train.parquet.gzip"
- config_name: nld_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/nld_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: bul_Cyrl-deu_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-deu_Latn/train.parquet.gzip"
- config_name: ben_Beng-pes_Arab
  data_files: 
  - split: train 
    path: "data/ben_Beng-pes_Arab/train.parquet.gzip"
- config_name: jpn_Jpan-sqi_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-sqi_Latn/train.parquet.gzip"
- config_name: ben_Beng-kor_Hang
  data_files: 
  - split: train 
    path: "data/ben_Beng-kor_Hang/train.parquet.gzip"
- config_name: ben_Beng-pan_Guru
  data_files: 
  - split: train 
    path: "data/ben_Beng-pan_Guru/train.parquet.gzip"
- config_name: hun_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-sqi_Latn/train.parquet.gzip"
- config_name: nld_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/nld_Latn-pan_Guru/train.parquet.gzip"
- config_name: kor_Hang-swh_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-swh_Latn/train.parquet.gzip"
- config_name: fra_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-ind_Latn/train.parquet.gzip"
- config_name: ell_Grek-sqi_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-sqi_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-kor_Hang
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-kor_Hang/train.parquet.gzip"
- config_name: tet_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/tet_Latn-zho_Hant/train.parquet.gzip"
- config_name: fra_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-nld_Latn/train.parquet.gzip"
- config_name: eng_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/eng_Latn-zho_Hant/train.parquet.gzip"
- config_name: ory_Orya-pol_Latn
  data_files: 
  - split: train 
    path: "data/ory_Orya-pol_Latn/train.parquet.gzip"
- config_name: plt_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/plt_Latn-npi_Deva/train.parquet.gzip"
- config_name: ben_Beng-zho_Hans
  data_files: 
  - split: train 
    path: "data/ben_Beng-zho_Hans/train.parquet.gzip"
- config_name: cat_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/cat_Latn-khm_Khmr/train.parquet.gzip"
- config_name: swe_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/swe_Latn-zho_Hant/train.parquet.gzip"
- config_name: arb_Arab-hin_Deva
  data_files: 
  - split: train 
    path: "data/arb_Arab-hin_Deva/train.parquet.gzip"
- config_name: pes_Arab-zho_Hans
  data_files: 
  - split: train 
    path: "data/pes_Arab-zho_Hans/train.parquet.gzip"
- config_name: ind_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/ind_Latn-zho_Hant/train.parquet.gzip"
- config_name: por_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/por_Latn-urd_Arab/train.parquet.gzip"
- config_name: khm_Khmr-por_Latn
  data_files: 
  - split: train 
    path: "data/khm_Khmr-por_Latn/train.parquet.gzip"
- config_name: heb_Hebr-pol_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-pol_Latn/train.parquet.gzip"
- config_name: ell_Grek-zho_Hant
  data_files: 
  - split: train 
    path: "data/ell_Grek-zho_Hant/train.parquet.gzip"
- config_name: por_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-swe_Latn/train.parquet.gzip"
- config_name: ben_Beng-ind_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-ind_Latn/train.parquet.gzip"
- config_name: spa_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/spa_Latn-urd_Arab/train.parquet.gzip"
- config_name: srp_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/srp_Latn-zho_Hans/train.parquet.gzip"
- config_name: deu_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/deu_Latn-kor_Hang/train.parquet.gzip"
- config_name: khm_Khmr-plt_Latn
  data_files: 
  - split: train 
    path: "data/khm_Khmr-plt_Latn/train.parquet.gzip"
- config_name: plt_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/plt_Latn-ory_Orya/train.parquet.gzip"
- config_name: bul_Cyrl-ell_Grek
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-ell_Grek/train.parquet.gzip"
- config_name: spa_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-hun_Latn/train.parquet.gzip"
- config_name: heb_Hebr-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/heb_Hebr-mkd_Cyrl/train.parquet.gzip"
- config_name: jpn_Jpan-pan_Guru
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-pan_Guru/train.parquet.gzip"
- config_name: tur_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/tur_Latn-zho_Hant/train.parquet.gzip"
- config_name: plt_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-sqi_Latn/train.parquet.gzip"
- config_name: fil_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-nld_Latn/train.parquet.gzip"
- config_name: arb_Arab-eng_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-eng_Latn/train.parquet.gzip"
- config_name: pol_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-ron_Latn/train.parquet.gzip"
- config_name: hun_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/hun_Latn-mya_Mymr/train.parquet.gzip"
- config_name: ell_Grek-npi_Deva
  data_files: 
  - split: train 
    path: "data/ell_Grek-npi_Deva/train.parquet.gzip"
- config_name: amh_Ethi-ind_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-ind_Latn/train.parquet.gzip"
- config_name: fil_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/fil_Latn-zho_Hant/train.parquet.gzip"
- config_name: arb_Arab-npi_Deva
  data_files: 
  - split: train 
    path: "data/arb_Arab-npi_Deva/train.parquet.gzip"
- config_name: fra_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/fra_Latn-urd_Arab/train.parquet.gzip"
- config_name: arb_Arab-yor_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-yor_Latn/train.parquet.gzip"
- config_name: mya_Mymr-zho_Hant
  data_files: 
  - split: train 
    path: "data/mya_Mymr-zho_Hant/train.parquet.gzip"
- config_name: ell_Grek-urd_Arab
  data_files: 
  - split: train 
    path: "data/ell_Grek-urd_Arab/train.parquet.gzip"
- config_name: ben_Beng-zho_Hant
  data_files: 
  - split: train 
    path: "data/ben_Beng-zho_Hant/train.parquet.gzip"
- config_name: arb_Arab-bul_Cyrl
  data_files: 
  - split: train 
    path: "data/arb_Arab-bul_Cyrl/train.parquet.gzip"
- config_name: ben_Beng-sqi_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-sqi_Latn/train.parquet.gzip"
- config_name: cat_Latn-pes_Arab
  data_files: 
  - split: train 
    path: "data/cat_Latn-pes_Arab/train.parquet.gzip"
- config_name: ayr_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-sqi_Latn/train.parquet.gzip"
- config_name: pan_Guru-swh_Latn
  data_files: 
  - split: train 
    path: "data/pan_Guru-swh_Latn/train.parquet.gzip"
- config_name: kor_Hang-ron_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-ron_Latn/train.parquet.gzip"
- config_name: spa_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/spa_Latn-zho_Hans/train.parquet.gzip"
- config_name: ayr_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-por_Latn/train.parquet.gzip"
- config_name: eng_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-swh_Latn/train.parquet.gzip"
- config_name: heb_Hebr-nld_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-nld_Latn/train.parquet.gzip"
- config_name: mya_Mymr-pol_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-pol_Latn/train.parquet.gzip"
- config_name: deu_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/deu_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: pol_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-swh_Latn/train.parquet.gzip"
- config_name: nld_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-pol_Latn/train.parquet.gzip"
- config_name: hun_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/hun_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: dan_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-ind_Latn/train.parquet.gzip"
- config_name: ayr_Latn-cat_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-cat_Latn/train.parquet.gzip"
- config_name: spa_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-ind_Latn/train.parquet.gzip"
- config_name: swe_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/swe_Latn-zho_Hans/train.parquet.gzip"
- config_name: deu_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-sqi_Latn/train.parquet.gzip"
- config_name: deu_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-fra_Latn/train.parquet.gzip"
- config_name: hun_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-plt_Latn/train.parquet.gzip"
- config_name: ces_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-pol_Latn/train.parquet.gzip"
- config_name: ell_Grek-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/ell_Grek-jpn_Jpan/train.parquet.gzip"
- config_name: fil_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-plt_Latn/train.parquet.gzip"
- config_name: plt_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-tur_Latn/train.parquet.gzip"
- config_name: pes_Arab-por_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-por_Latn/train.parquet.gzip"
- config_name: fil_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-swh_Latn/train.parquet.gzip"
- config_name: hin_Deva-pol_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-pol_Latn/train.parquet.gzip"
- config_name: mya_Mymr-zho_Hans
  data_files: 
  - split: train 
    path: "data/mya_Mymr-zho_Hans/train.parquet.gzip"
- config_name: ind_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-tet_Latn/train.parquet.gzip"
- config_name: swe_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/swe_Latn-swh_Latn/train.parquet.gzip"
- config_name: ayr_Latn-ben_Beng
  data_files: 
  - split: train 
    path: "data/ayr_Latn-ben_Beng/train.parquet.gzip"
- config_name: ces_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/ces_Latn-ory_Orya/train.parquet.gzip"
- config_name: ita_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/ita_Latn-zho_Hant/train.parquet.gzip"
- config_name: jpn_Jpan-nld_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-nld_Latn/train.parquet.gzip"
- config_name: fra_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-sqi_Latn/train.parquet.gzip"
- config_name: ces_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/ces_Latn-pan_Guru/train.parquet.gzip"
- config_name: ita_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/ita_Latn-mya_Mymr/train.parquet.gzip"
- config_name: mya_Mymr-swe_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-swe_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-zho_Hans
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-zho_Hans/train.parquet.gzip"
- config_name: spa_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/spa_Latn-zho_Hant/train.parquet.gzip"
- config_name: ind_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-ron_Latn/train.parquet.gzip"
- config_name: swh_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/swh_Latn-yor_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-swh_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-swh_Latn/train.parquet.gzip"
- config_name: srp_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/srp_Latn-swh_Latn/train.parquet.gzip"
- config_name: dan_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-hun_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-pol_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-pol_Latn/train.parquet.gzip"
- config_name: deu_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-tet_Latn/train.parquet.gzip"
- config_name: heb_Hebr-plt_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-plt_Latn/train.parquet.gzip"
- config_name: ita_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-pol_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-mya_Mymr
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-mya_Mymr/train.parquet.gzip"
- config_name: plt_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-tet_Latn/train.parquet.gzip"
- config_name: hun_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-srp_Latn/train.parquet.gzip"
- config_name: cat_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/cat_Latn-ory_Orya/train.parquet.gzip"
- config_name: ita_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-sqi_Latn/train.parquet.gzip"
- config_name: kor_Hang-zho_Hant
  data_files: 
  - split: train 
    path: "data/kor_Hang-zho_Hant/train.parquet.gzip"
- config_name: kor_Hang-pan_Guru
  data_files: 
  - split: train 
    path: "data/kor_Hang-pan_Guru/train.parquet.gzip"
- config_name: spa_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-por_Latn/train.parquet.gzip"
- config_name: ben_Beng-fil_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-fil_Latn/train.parquet.gzip"
- config_name: ayr_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/ayr_Latn-zho_Hans/train.parquet.gzip"
- config_name: fra_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/fra_Latn-zho_Hans/train.parquet.gzip"
- config_name: spa_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-swh_Latn/train.parquet.gzip"
- config_name: eng_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/eng_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: fil_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-ita_Latn/train.parquet.gzip"
- config_name: ayr_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-fra_Latn/train.parquet.gzip"
- config_name: ben_Beng-hun_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-hun_Latn/train.parquet.gzip"
- config_name: mya_Mymr-urd_Arab
  data_files: 
  - split: train 
    path: "data/mya_Mymr-urd_Arab/train.parquet.gzip"
- config_name: cat_Latn-deu_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-deu_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-plt_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-plt_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-spa_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-spa_Latn/train.parquet.gzip"
- config_name: fra_Latn-ewe_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-ewe_Latn/train.parquet.gzip"
- config_name: deu_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/deu_Latn-zho_Hans/train.parquet.gzip"
- config_name: spa_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/spa_Latn-ory_Orya/train.parquet.gzip"
- config_name: deu_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-swh_Latn/train.parquet.gzip"
- config_name: eng_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-srp_Latn/train.parquet.gzip"
- config_name: ind_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/ind_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: plt_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-swh_Latn/train.parquet.gzip"
- config_name: hun_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/hun_Latn-zho_Hans/train.parquet.gzip"
- config_name: ell_Grek-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/ell_Grek-mkd_Cyrl/train.parquet.gzip"
- config_name: eng_Latn-zul_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-zul_Latn/train.parquet.gzip"
- config_name: ayr_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/ayr_Latn-kor_Hang/train.parquet.gzip"
- config_name: spa_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/spa_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: ces_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/ces_Latn-kor_Hang/train.parquet.gzip"
- config_name: ind_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/ind_Latn-zho_Hans/train.parquet.gzip"
- config_name: kor_Hang-plt_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-plt_Latn/train.parquet.gzip"
- config_name: arb_Arab-ory_Orya
  data_files: 
  - split: train 
    path: "data/arb_Arab-ory_Orya/train.parquet.gzip"
- config_name: eng_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-hun_Latn/train.parquet.gzip"
- config_name: deu_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-pol_Latn/train.parquet.gzip"
- config_name: tur_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/tur_Latn-urd_Arab/train.parquet.gzip"
- config_name: ayr_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ayr_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: mkd_Cyrl-ron_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-ron_Latn/train.parquet.gzip"
- config_name: fra_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-plt_Latn/train.parquet.gzip"
- config_name: ben_Beng-mya_Mymr
  data_files: 
  - split: train 
    path: "data/ben_Beng-mya_Mymr/train.parquet.gzip"
- config_name: sqi_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/sqi_Latn-swe_Latn/train.parquet.gzip"
- config_name: arb_Arab-por_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-por_Latn/train.parquet.gzip"
- config_name: deu_Latn-heb_Hebr
  data_files: 
  - split: train 
    path: "data/deu_Latn-heb_Hebr/train.parquet.gzip"
- config_name: ron_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ron_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: hun_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/hun_Latn-khm_Khmr/train.parquet.gzip"
- config_name: por_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/por_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: pes_Arab-fil_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-fil_Latn/train.parquet.gzip"
- config_name: hun_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-pol_Latn/train.parquet.gzip"
- config_name: amh_Ethi-spa_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-spa_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-srp_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-srp_Latn/train.parquet.gzip"
- config_name: pes_Arab-ron_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-ron_Latn/train.parquet.gzip"
- config_name: amh_Ethi-por_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-por_Latn/train.parquet.gzip"
- config_name: ben_Beng-ces_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-ces_Latn/train.parquet.gzip"
- config_name: hin_Deva-sqi_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-sqi_Latn/train.parquet.gzip"
- config_name: khm_Khmr-nld_Latn
  data_files: 
  - split: train 
    path: "data/khm_Khmr-nld_Latn/train.parquet.gzip"
- config_name: sqi_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/sqi_Latn-zho_Hans/train.parquet.gzip"
- config_name: fra_Latn-ckb_Arab
  data_files: 
  - split: train 
    path: "data/fra_Latn-ckb_Arab/train.parquet.gzip"
- config_name: eng_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-yor_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-tur_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-tur_Latn/train.parquet.gzip"
- config_name: ron_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/ron_Latn-srp_Latn/train.parquet.gzip"
- config_name: ckb_Arab-ron_Latn
  data_files: 
  - split: train 
    path: "data/ckb_Arab-ron_Latn/train.parquet.gzip"
- config_name: ayr_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/ayr_Latn-khm_Khmr/train.parquet.gzip"
- config_name: eng_Latn-heb_Hebr
  data_files: 
  - split: train 
    path: "data/eng_Latn-heb_Hebr/train.parquet.gzip"
- config_name: hun_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/hun_Latn-urd_Arab/train.parquet.gzip"
- config_name: arb_Arab-ayr_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-ayr_Latn/train.parquet.gzip"
- config_name: ron_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/ron_Latn-urd_Arab/train.parquet.gzip"
- config_name: pol_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/pol_Latn-zho_Hans/train.parquet.gzip"
- config_name: arb_Arab-fra_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-fra_Latn/train.parquet.gzip"
- config_name: hin_Deva-ory_Orya
  data_files: 
  - split: train 
    path: "data/hin_Deva-ory_Orya/train.parquet.gzip"
- config_name: fra_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/fra_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: plt_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/plt_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: arb_Arab-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/arb_Arab-rus_Cyrl/train.parquet.gzip"
- config_name: bul_Cyrl-ron_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-ron_Latn/train.parquet.gzip"
- config_name: ben_Beng-pol_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-pol_Latn/train.parquet.gzip"
- config_name: ita_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/ita_Latn-zho_Hans/train.parquet.gzip"
- config_name: fra_Latn-bam_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-bam_Latn/train.parquet.gzip"
- config_name: hin_Deva-plt_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-plt_Latn/train.parquet.gzip"
- config_name: ell_Grek-por_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-por_Latn/train.parquet.gzip"
- config_name: ayr_Latn-spa_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-spa_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-ces_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-ces_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-ben_Beng
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-ben_Beng/train.parquet.gzip"
- config_name: bul_Cyrl-ind_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-ind_Latn/train.parquet.gzip"
- config_name: fil_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-srp_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-urd_Arab
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-urd_Arab/train.parquet.gzip"
- config_name: spa_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-nld_Latn/train.parquet.gzip"
- config_name: ita_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/ita_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: cat_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/cat_Latn-zho_Hans/train.parquet.gzip"
- config_name: eng_Latn-luo_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-luo_Latn/train.parquet.gzip"
- config_name: amh_Ethi-zho_Hant
  data_files: 
  - split: train 
    path: "data/amh_Ethi-zho_Hant/train.parquet.gzip"
- config_name: ben_Beng-plt_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-plt_Latn/train.parquet.gzip"
- config_name: cat_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-ita_Latn/train.parquet.gzip"
- config_name: nld_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-tur_Latn/train.parquet.gzip"
- config_name: ron_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/ron_Latn-zho_Hant/train.parquet.gzip"
- config_name: amh_Ethi-srp_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-srp_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-mkd_Cyrl/train.parquet.gzip"
- config_name: ces_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-sqi_Latn/train.parquet.gzip"
- config_name: amh_Ethi-eng_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-eng_Latn/train.parquet.gzip"
- config_name: pes_Arab-ind_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-ind_Latn/train.parquet.gzip"
- config_name: ayr_Latn-pes_Arab
  data_files: 
  - split: train 
    path: "data/ayr_Latn-pes_Arab/train.parquet.gzip"
- config_name: eng_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/eng_Latn-urd_Arab/train.parquet.gzip"
- config_name: cat_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-pol_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-tur_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-tur_Latn/train.parquet.gzip"
- config_name: eng_Latn-fil_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-fil_Latn/train.parquet.gzip"
- config_name: arb_Arab-pol_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-pol_Latn/train.parquet.gzip"
- config_name: rus_Cyrl-sqi_Latn
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-sqi_Latn/train.parquet.gzip"
- config_name: dan_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/dan_Latn-mya_Mymr/train.parquet.gzip"
- config_name: cat_Latn-spa_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-spa_Latn/train.parquet.gzip"
- config_name: fra_Latn-bbj_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-bbj_Latn/train.parquet.gzip"
- config_name: fil_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/fil_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: swe_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/swe_Latn-tur_Latn/train.parquet.gzip"
- config_name: spa_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/spa_Latn-kor_Hang/train.parquet.gzip"
- config_name: cat_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-ind_Latn/train.parquet.gzip"
- config_name: amh_Ethi-plt_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-plt_Latn/train.parquet.gzip"
- config_name: ben_Beng-por_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-por_Latn/train.parquet.gzip"
- config_name: dan_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-pol_Latn/train.parquet.gzip"
- config_name: deu_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/deu_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: bul_Cyrl-eng_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-eng_Latn/train.parquet.gzip"
- config_name: pes_Arab-npi_Deva
  data_files: 
  - split: train 
    path: "data/pes_Arab-npi_Deva/train.parquet.gzip"
- config_name: heb_Hebr-tur_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-tur_Latn/train.parquet.gzip"
- config_name: ell_Grek-ita_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-ita_Latn/train.parquet.gzip"
- config_name: ben_Beng-swe_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-swe_Latn/train.parquet.gzip"
- config_name: spa_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/spa_Latn-npi_Deva/train.parquet.gzip"
- config_name: ind_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/ind_Latn-urd_Arab/train.parquet.gzip"
- config_name: ben_Beng-eng_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-eng_Latn/train.parquet.gzip"
- config_name: pes_Arab-hun_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-hun_Latn/train.parquet.gzip"
- config_name: ron_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/ron_Latn-tur_Latn/train.parquet.gzip"
- config_name: cat_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/cat_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: kor_Hang-pol_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-pol_Latn/train.parquet.gzip"
- config_name: npi_Deva-tur_Latn
  data_files: 
  - split: train 
    path: "data/npi_Deva-tur_Latn/train.parquet.gzip"
- config_name: hin_Deva-urd_Arab
  data_files: 
  - split: train 
    path: "data/hin_Deva-urd_Arab/train.parquet.gzip"
- config_name: pes_Arab-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/pes_Arab-mkd_Cyrl/train.parquet.gzip"
- config_name: rus_Cyrl-srp_Latn
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-srp_Latn/train.parquet.gzip"
- config_name: rus_Cyrl-tet_Latn
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-tet_Latn/train.parquet.gzip"
- config_name: fil_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-tur_Latn/train.parquet.gzip"
- config_name: hin_Deva-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/hin_Deva-jpn_Jpan/train.parquet.gzip"
- config_name: eng_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/eng_Latn-kor_Hang/train.parquet.gzip"
- config_name: amh_Ethi-swh_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-swh_Latn/train.parquet.gzip"
- config_name: fra_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-ron_Latn/train.parquet.gzip"
- config_name: arb_Arab-fil_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-fil_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-nld_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-nld_Latn/train.parquet.gzip"
- config_name: sqi_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/sqi_Latn-zho_Hant/train.parquet.gzip"
- config_name: deu_Latn-hin_Deva
  data_files: 
  - split: train 
    path: "data/deu_Latn-hin_Deva/train.parquet.gzip"
- config_name: arb_Arab-spa_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-spa_Latn/train.parquet.gzip"
- config_name: fra_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-por_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-ron_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-ron_Latn/train.parquet.gzip"
- config_name: eng_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-plt_Latn/train.parquet.gzip"
- config_name: plt_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/plt_Latn-mya_Mymr/train.parquet.gzip"
- config_name: spa_Latn-pes_Arab
  data_files: 
  - split: train 
    path: "data/spa_Latn-pes_Arab/train.parquet.gzip"
- config_name: hin_Deva-swe_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-swe_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-zho_Hans
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-zho_Hans/train.parquet.gzip"
- config_name: eng_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-sqi_Latn/train.parquet.gzip"
- config_name: plt_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/plt_Latn-zho_Hant/train.parquet.gzip"
- config_name: nld_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-swh_Latn/train.parquet.gzip"
- config_name: spa_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-tet_Latn/train.parquet.gzip"
- config_name: deu_Latn-fil_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-fil_Latn/train.parquet.gzip"
- config_name: ayr_Latn-dan_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-dan_Latn/train.parquet.gzip"
- config_name: arb_Arab-cat_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-cat_Latn/train.parquet.gzip"
- config_name: ayr_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/ayr_Latn-ory_Orya/train.parquet.gzip"
- config_name: ayr_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-nld_Latn/train.parquet.gzip"
- config_name: dan_Latn-ell_Grek
  data_files: 
  - split: train 
    path: "data/dan_Latn-ell_Grek/train.parquet.gzip"
- config_name: nld_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-srp_Latn/train.parquet.gzip"
- config_name: ces_Latn-dan_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-dan_Latn/train.parquet.gzip"
- config_name: npi_Deva-zho_Hans
  data_files: 
  - split: train 
    path: "data/npi_Deva-zho_Hans/train.parquet.gzip"
- config_name: ces_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-swh_Latn/train.parquet.gzip"
- config_name: spa_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-ita_Latn/train.parquet.gzip"
- config_name: cat_Latn-dan_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-dan_Latn/train.parquet.gzip"
- config_name: cat_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-sqi_Latn/train.parquet.gzip"
- config_name: ind_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-srp_Latn/train.parquet.gzip"
- config_name: por_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/por_Latn-zho_Hant/train.parquet.gzip"
- config_name: ita_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-plt_Latn/train.parquet.gzip"
- config_name: eng_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-por_Latn/train.parquet.gzip"
- config_name: arb_Arab-khm_Khmr
  data_files: 
  - split: train 
    path: "data/arb_Arab-khm_Khmr/train.parquet.gzip"
- config_name: eng_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-nld_Latn/train.parquet.gzip"
- config_name: rus_Cyrl-yor_Latn
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-yor_Latn/train.parquet.gzip"
- config_name: ces_Latn-spa_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-spa_Latn/train.parquet.gzip"
- config_name: dan_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-tur_Latn/train.parquet.gzip"
- config_name: pol_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/pol_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: ayr_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-swe_Latn/train.parquet.gzip"
- config_name: fil_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-pol_Latn/train.parquet.gzip"
- config_name: eng_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/eng_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: dan_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/dan_Latn-khm_Khmr/train.parquet.gzip"
- config_name: hun_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/hun_Latn-pan_Guru/train.parquet.gzip"
- config_name: ell_Grek-yor_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-yor_Latn/train.parquet.gzip"
- config_name: spa_Latn-fil_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-fil_Latn/train.parquet.gzip"
- config_name: ind_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-sqi_Latn/train.parquet.gzip"
- config_name: ell_Grek-spa_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-spa_Latn/train.parquet.gzip"
- config_name: hin_Deva-tur_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-tur_Latn/train.parquet.gzip"
- config_name: sqi_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/sqi_Latn-swh_Latn/train.parquet.gzip"
- config_name: eng_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-pol_Latn/train.parquet.gzip"
- config_name: ayr_Latn-deu_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-deu_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-fra_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-fra_Latn/train.parquet.gzip"
- config_name: deu_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/deu_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: ckb_Arab-por_Latn
  data_files: 
  - split: train 
    path: "data/ckb_Arab-por_Latn/train.parquet.gzip"
- config_name: kor_Hang-mya_Mymr
  data_files: 
  - split: train 
    path: "data/kor_Hang-mya_Mymr/train.parquet.gzip"
- config_name: nld_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-ron_Latn/train.parquet.gzip"
- config_name: eng_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/eng_Latn-ory_Orya/train.parquet.gzip"
- config_name: ind_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-tur_Latn/train.parquet.gzip"
- config_name: nld_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/nld_Latn-urd_Arab/train.parquet.gzip"
- config_name: pes_Arab-ita_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-ita_Latn/train.parquet.gzip"
- config_name: npi_Deva-pan_Guru
  data_files: 
  - split: train 
    path: "data/npi_Deva-pan_Guru/train.parquet.gzip"
- config_name: cat_Latn-ces_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-ces_Latn/train.parquet.gzip"
- config_name: eng_Latn-ind_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-ind_Latn/train.parquet.gzip"
- config_name: ind_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/ind_Latn-npi_Deva/train.parquet.gzip"
- config_name: eng_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-swe_Latn/train.parquet.gzip"
- config_name: dan_Latn-spa_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-spa_Latn/train.parquet.gzip"
- config_name: ell_Grek-pol_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-pol_Latn/train.parquet.gzip"
- config_name: plt_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/plt_Latn-urd_Arab/train.parquet.gzip"
- config_name: arb_Arab-urd_Arab
  data_files: 
  - split: train 
    path: "data/arb_Arab-urd_Arab/train.parquet.gzip"
- config_name: heb_Hebr-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/heb_Hebr-rus_Cyrl/train.parquet.gzip"
- config_name: spa_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/spa_Latn-khm_Khmr/train.parquet.gzip"
- config_name: ayr_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-srp_Latn/train.parquet.gzip"
- config_name: fra_Latn-fon_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-fon_Latn/train.parquet.gzip"
- config_name: pes_Arab-fra_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-fra_Latn/train.parquet.gzip"
- config_name: hin_Deva-ind_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-ind_Latn/train.parquet.gzip"
- config_name: ben_Beng-deu_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-deu_Latn/train.parquet.gzip"
- config_name: spa_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-pol_Latn/train.parquet.gzip"
- config_name: nld_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/nld_Latn-por_Latn/train.parquet.gzip"
- config_name: ckb_Arab-pol_Latn
  data_files: 
  - split: train 
    path: "data/ckb_Arab-pol_Latn/train.parquet.gzip"
- config_name: ayr_Latn-ces_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-ces_Latn/train.parquet.gzip"
- config_name: cat_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/cat_Latn-urd_Arab/train.parquet.gzip"
- config_name: fra_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/fra_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: fra_Latn-heb_Hebr
  data_files: 
  - split: train 
    path: "data/fra_Latn-heb_Hebr/train.parquet.gzip"
- config_name: fra_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-swh_Latn/train.parquet.gzip"
- config_name: pes_Arab-ory_Orya
  data_files: 
  - split: train 
    path: "data/pes_Arab-ory_Orya/train.parquet.gzip"
- config_name: dan_Latn-deu_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-deu_Latn/train.parquet.gzip"
- config_name: arb_Arab-dan_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-dan_Latn/train.parquet.gzip"
- config_name: ces_Latn-deu_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-deu_Latn/train.parquet.gzip"
- config_name: deu_Latn-ell_Grek
  data_files: 
  - split: train 
    path: "data/deu_Latn-ell_Grek/train.parquet.gzip"
- config_name: hun_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/hun_Latn-kor_Hang/train.parquet.gzip"
- config_name: fra_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/fra_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: ell_Grek-swh_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-swh_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-plt_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-plt_Latn/train.parquet.gzip"
- config_name: spa_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-plt_Latn/train.parquet.gzip"
- config_name: ell_Grek-plt_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-plt_Latn/train.parquet.gzip"
- config_name: spa_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/spa_Latn-mya_Mymr/train.parquet.gzip"
- config_name: arb_Arab-ell_Grek
  data_files: 
  - split: train 
    path: "data/arb_Arab-ell_Grek/train.parquet.gzip"
- config_name: ell_Grek-mya_Mymr
  data_files: 
  - split: train 
    path: "data/ell_Grek-mya_Mymr/train.parquet.gzip"
- config_name: pes_Arab-swh_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-swh_Latn/train.parquet.gzip"
- config_name: heb_Hebr-swe_Latn
  data_files: 
  - split: train 
    path: "data/heb_Hebr-swe_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-rus_Cyrl/train.parquet.gzip"
- config_name: eng_Latn-ckb_Arab
  data_files: 
  - split: train 
    path: "data/eng_Latn-ckb_Arab/train.parquet.gzip"
- config_name: arb_Arab-swh_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-swh_Latn/train.parquet.gzip"
- config_name: cat_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/cat_Latn-kor_Hang/train.parquet.gzip"
- config_name: cat_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/cat_Latn-zho_Hant/train.parquet.gzip"
- config_name: eng_Latn-spa_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-spa_Latn/train.parquet.gzip"
- config_name: amh_Ethi-tur_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-tur_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-mkd_Cyrl/train.parquet.gzip"
- config_name: amh_Ethi-arb_Arab
  data_files: 
  - split: train 
    path: "data/amh_Ethi-arb_Arab/train.parquet.gzip"
- config_name: pes_Arab-swe_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-swe_Latn/train.parquet.gzip"
- config_name: dan_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-swh_Latn/train.parquet.gzip"
- config_name: hin_Deva-nld_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-nld_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-cat_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-cat_Latn/train.parquet.gzip"
- config_name: ces_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-por_Latn/train.parquet.gzip"
- config_name: arb_Arab-ces_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-ces_Latn/train.parquet.gzip"
- config_name: eng_Latn-lit_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-lit_Latn/train.parquet.gzip"
- config_name: spa_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-tur_Latn/train.parquet.gzip"
- config_name: fil_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-swe_Latn/train.parquet.gzip"
- config_name: fra_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-ita_Latn/train.parquet.gzip"
- config_name: ben_Beng-fra_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-fra_Latn/train.parquet.gzip"
- config_name: pol_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-swe_Latn/train.parquet.gzip"
- config_name: ces_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-ita_Latn/train.parquet.gzip"
- config_name: deu_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/deu_Latn-zho_Hant/train.parquet.gzip"
- config_name: ind_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/ind_Latn-kor_Hang/train.parquet.gzip"
- config_name: ell_Grek-fil_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-fil_Latn/train.parquet.gzip"
- config_name: deu_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-swe_Latn/train.parquet.gzip"
- config_name: arb_Arab-nld_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-nld_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-swe_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-swe_Latn/train.parquet.gzip"
- config_name: hin_Deva-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/hin_Deva-mkd_Cyrl/train.parquet.gzip"
- config_name: fra_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-srp_Latn/train.parquet.gzip"
- config_name: kor_Hang-srp_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-srp_Latn/train.parquet.gzip"
- config_name: amh_Ethi-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/amh_Ethi-rus_Cyrl/train.parquet.gzip"
- config_name: cat_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-swe_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-srp_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-srp_Latn/train.parquet.gzip"
- config_name: deu_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/deu_Latn-npi_Deva/train.parquet.gzip"
- config_name: pes_Arab-heb_Hebr
  data_files: 
  - split: train 
    path: "data/pes_Arab-heb_Hebr/train.parquet.gzip"
- config_name: dan_Latn-eng_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-eng_Latn/train.parquet.gzip"
- config_name: arb_Arab-heb_Hebr
  data_files: 
  - split: train 
    path: "data/arb_Arab-heb_Hebr/train.parquet.gzip"
- config_name: mya_Mymr-ron_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-ron_Latn/train.parquet.gzip"
- config_name: cat_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/cat_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: mkd_Cyrl-npi_Deva
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-npi_Deva/train.parquet.gzip"
- config_name: amh_Ethi-ayr_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-ayr_Latn/train.parquet.gzip"
- config_name: ita_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-yor_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-zho_Hans
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-zho_Hans/train.parquet.gzip"
- config_name: arb_Arab-ckb_Arab
  data_files: 
  - split: train 
    path: "data/arb_Arab-ckb_Arab/train.parquet.gzip"
- config_name: ces_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/ces_Latn-zho_Hans/train.parquet.gzip"
- config_name: dan_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-swe_Latn/train.parquet.gzip"
- config_name: plt_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/plt_Latn-zho_Hans/train.parquet.gzip"
- config_name: hun_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-por_Latn/train.parquet.gzip"
- config_name: hin_Deva-hun_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-hun_Latn/train.parquet.gzip"
- config_name: pan_Guru-urd_Arab
  data_files: 
  - split: train 
    path: "data/pan_Guru-urd_Arab/train.parquet.gzip"
- config_name: mkd_Cyrl-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-rus_Cyrl/train.parquet.gzip"
- config_name: jpn_Jpan-npi_Deva
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-npi_Deva/train.parquet.gzip"
- config_name: cat_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-fra_Latn/train.parquet.gzip"
- config_name: cat_Latn-heb_Hebr
  data_files: 
  - split: train 
    path: "data/cat_Latn-heb_Hebr/train.parquet.gzip"
- config_name: ell_Grek-tet_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-tet_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-mya_Mymr
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-mya_Mymr/train.parquet.gzip"
- config_name: dan_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-sqi_Latn/train.parquet.gzip"
- config_name: ell_Grek-pan_Guru
  data_files: 
  - split: train 
    path: "data/ell_Grek-pan_Guru/train.parquet.gzip"
- config_name: ben_Beng-ell_Grek
  data_files: 
  - split: train 
    path: "data/ben_Beng-ell_Grek/train.parquet.gzip"
- config_name: fil_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/fil_Latn-khm_Khmr/train.parquet.gzip"
- config_name: arb_Arab-sqi_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-sqi_Latn/train.parquet.gzip"
- config_name: pol_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-yor_Latn/train.parquet.gzip"
- config_name: fil_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/fil_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: ces_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-tet_Latn/train.parquet.gzip"
- config_name: deu_Latn-spa_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-spa_Latn/train.parquet.gzip"
- config_name: deu_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-hun_Latn/train.parquet.gzip"
- config_name: ita_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-por_Latn/train.parquet.gzip"
- config_name: pes_Arab-zho_Hant
  data_files: 
  - split: train 
    path: "data/pes_Arab-zho_Hant/train.parquet.gzip"
- config_name: mkd_Cyrl-pol_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-pol_Latn/train.parquet.gzip"
- config_name: fra_Latn-hun_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-hun_Latn/train.parquet.gzip"
- config_name: eng_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/eng_Latn-pan_Guru/train.parquet.gzip"
- config_name: bul_Cyrl-sqi_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-sqi_Latn/train.parquet.gzip"
- config_name: hin_Deva-zho_Hans
  data_files: 
  - split: train 
    path: "data/hin_Deva-zho_Hans/train.parquet.gzip"
- config_name: sqi_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/sqi_Latn-urd_Arab/train.parquet.gzip"
- config_name: eng_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-ron_Latn/train.parquet.gzip"
- config_name: spa_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-ron_Latn/train.parquet.gzip"
- config_name: ell_Grek-ron_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-ron_Latn/train.parquet.gzip"
- config_name: dan_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/dan_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: pes_Arab-khm_Khmr
  data_files: 
  - split: train 
    path: "data/pes_Arab-khm_Khmr/train.parquet.gzip"
- config_name: hin_Deva-por_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-por_Latn/train.parquet.gzip"
- config_name: arb_Arab-ron_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-ron_Latn/train.parquet.gzip"
- config_name: hun_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/hun_Latn-zho_Hant/train.parquet.gzip"
- config_name: bul_Cyrl-por_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-por_Latn/train.parquet.gzip"
- config_name: pes_Arab-plt_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-plt_Latn/train.parquet.gzip"
- config_name: fra_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-pol_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-zho_Hant
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-zho_Hant/train.parquet.gzip"
- config_name: kor_Hang-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/kor_Hang-rus_Cyrl/train.parquet.gzip"
- config_name: pan_Guru-por_Latn
  data_files: 
  - split: train 
    path: "data/pan_Guru-por_Latn/train.parquet.gzip"
- config_name: nld_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/nld_Latn-ory_Orya/train.parquet.gzip"
- config_name: ell_Grek-fra_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-fra_Latn/train.parquet.gzip"
- config_name: pol_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-por_Latn/train.parquet.gzip"
- config_name: fra_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-tur_Latn/train.parquet.gzip"
- config_name: fra_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/fra_Latn-npi_Deva/train.parquet.gzip"
- config_name: dan_Latn-heb_Hebr
  data_files: 
  - split: train 
    path: "data/dan_Latn-heb_Hebr/train.parquet.gzip"
- config_name: eng_Latn-twi_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-twi_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-npi_Deva
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-npi_Deva/train.parquet.gzip"
- config_name: mya_Mymr-swh_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-swh_Latn/train.parquet.gzip"
- config_name: ben_Beng-npi_Deva
  data_files: 
  - split: train 
    path: "data/ben_Beng-npi_Deva/train.parquet.gzip"
- config_name: kor_Hang-urd_Arab
  data_files: 
  - split: train 
    path: "data/kor_Hang-urd_Arab/train.parquet.gzip"
- config_name: amh_Ethi-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/amh_Ethi-jpn_Jpan/train.parquet.gzip"
- config_name: tur_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/tur_Latn-zho_Hans/train.parquet.gzip"
- config_name: swh_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/swh_Latn-zho_Hant/train.parquet.gzip"
- config_name: eng_Latn-lav_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-lav_Latn/train.parquet.gzip"
- config_name: deu_Latn-pes_Arab
  data_files: 
  - split: train 
    path: "data/deu_Latn-pes_Arab/train.parquet.gzip"
- config_name: deu_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-tur_Latn/train.parquet.gzip"
- config_name: eng_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-fra_Latn/train.parquet.gzip"
- config_name: eng_Latn-hau_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-hau_Latn/train.parquet.gzip"
- config_name: ell_Grek-kor_Hang
  data_files: 
  - split: train 
    path: "data/ell_Grek-kor_Hang/train.parquet.gzip"
- config_name: cat_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/cat_Latn-pan_Guru/train.parquet.gzip"
- config_name: ell_Grek-ory_Orya
  data_files: 
  - split: train 
    path: "data/ell_Grek-ory_Orya/train.parquet.gzip"
- config_name: ces_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-fra_Latn/train.parquet.gzip"
- config_name: arb_Arab-ben_Beng
  data_files: 
  - split: train 
    path: "data/arb_Arab-ben_Beng/train.parquet.gzip"
- config_name: hun_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-swe_Latn/train.parquet.gzip"
- config_name: jpn_Jpan-pol_Latn
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-pol_Latn/train.parquet.gzip"
- config_name: hun_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-tur_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-sqi_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-sqi_Latn/train.parquet.gzip"
- config_name: ita_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-nld_Latn/train.parquet.gzip"
- config_name: khm_Khmr-srp_Latn
  data_files: 
  - split: train 
    path: "data/khm_Khmr-srp_Latn/train.parquet.gzip"
- config_name: kor_Hang-npi_Deva
  data_files: 
  - split: train 
    path: "data/kor_Hang-npi_Deva/train.parquet.gzip"
- config_name: ayr_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/ayr_Latn-mya_Mymr/train.parquet.gzip"
- config_name: eng_Latn-est_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-est_Latn/train.parquet.gzip"
- config_name: spa_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-srp_Latn/train.parquet.gzip"
- config_name: ita_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/ita_Latn-swh_Latn/train.parquet.gzip"
- config_name: fra_Latn-wol_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-wol_Latn/train.parquet.gzip"
- config_name: hin_Deva-srp_Latn
  data_files: 
  - split: train 
    path: "data/hin_Deva-srp_Latn/train.parquet.gzip"
- config_name: npi_Deva-urd_Arab
  data_files: 
  - split: train 
    path: "data/npi_Deva-urd_Arab/train.parquet.gzip"
- config_name: npi_Deva-por_Latn
  data_files: 
  - split: train 
    path: "data/npi_Deva-por_Latn/train.parquet.gzip"
- config_name: dan_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/dan_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: dan_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/dan_Latn-zho_Hans/train.parquet.gzip"
- config_name: dan_Latn-zho_Hant
  data_files: 
  - split: train 
    path: "data/dan_Latn-zho_Hant/train.parquet.gzip"
- config_name: pol_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-sqi_Latn/train.parquet.gzip"
- config_name: ces_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-ron_Latn/train.parquet.gzip"
- config_name: cat_Latn-eng_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-eng_Latn/train.parquet.gzip"
- config_name: eng_Latn-pes_Arab
  data_files: 
  - split: train 
    path: "data/eng_Latn-pes_Arab/train.parquet.gzip"
- config_name: bul_Cyrl-tur_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-tur_Latn/train.parquet.gzip"
- config_name: ita_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/ita_Latn-pan_Guru/train.parquet.gzip"
- config_name: pes_Arab-mya_Mymr
  data_files: 
  - split: train 
    path: "data/pes_Arab-mya_Mymr/train.parquet.gzip"
- config_name: mkd_Cyrl-swh_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-swh_Latn/train.parquet.gzip"
- config_name: ces_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/ces_Latn-npi_Deva/train.parquet.gzip"
- config_name: hun_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-ita_Latn/train.parquet.gzip"
- config_name: ces_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/ces_Latn-mya_Mymr/train.parquet.gzip"
- config_name: fil_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/fil_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: fra_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/fra_Latn-khm_Khmr/train.parquet.gzip"
- config_name: khm_Khmr-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/khm_Khmr-mkd_Cyrl/train.parquet.gzip"
- config_name: ces_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/ces_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: fra_Latn-ory_Orya
  data_files: 
  - split: train 
    path: "data/fra_Latn-ory_Orya/train.parquet.gzip"
- config_name: ell_Grek-hin_Deva
  data_files: 
  - split: train 
    path: "data/ell_Grek-hin_Deva/train.parquet.gzip"
- config_name: ben_Beng-dan_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-dan_Latn/train.parquet.gzip"
- config_name: fil_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-fra_Latn/train.parquet.gzip"
- config_name: pes_Arab-kor_Hang
  data_files: 
  - split: train 
    path: "data/pes_Arab-kor_Hang/train.parquet.gzip"
- config_name: ind_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/ind_Latn-mya_Mymr/train.parquet.gzip"
- config_name: eng_Latn-kaz_Cyrl
  data_files: 
  - split: train 
    path: "data/eng_Latn-kaz_Cyrl/train.parquet.gzip"
- config_name: heb_Hebr-zho_Hans
  data_files: 
  - split: train 
    path: "data/heb_Hebr-zho_Hans/train.parquet.gzip"
- config_name: jpn_Jpan-zho_Hant
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-zho_Hant/train.parquet.gzip"
- config_name: pes_Arab-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/pes_Arab-jpn_Jpan/train.parquet.gzip"
- config_name: por_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-swh_Latn/train.parquet.gzip"
- config_name: cat_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-ron_Latn/train.parquet.gzip"
- config_name: ind_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-nld_Latn/train.parquet.gzip"
- config_name: plt_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-srp_Latn/train.parquet.gzip"
- config_name: amh_Ethi-kor_Hang
  data_files: 
  - split: train 
    path: "data/amh_Ethi-kor_Hang/train.parquet.gzip"
- config_name: pes_Arab-hin_Deva
  data_files: 
  - split: train 
    path: "data/pes_Arab-hin_Deva/train.parquet.gzip"
- config_name: por_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/por_Latn-zho_Hans/train.parquet.gzip"
- config_name: ell_Grek-khm_Khmr
  data_files: 
  - split: train 
    path: "data/ell_Grek-khm_Khmr/train.parquet.gzip"
- config_name: dan_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-srp_Latn/train.parquet.gzip"
- config_name: eng_Latn-guj_Gujr
  data_files: 
  - split: train 
    path: "data/eng_Latn-guj_Gujr/train.parquet.gzip"
- config_name: jpn_Jpan-ckb_Arab
  data_files: 
  - split: train 
    path: "data/jpn_Jpan-ckb_Arab/train.parquet.gzip"
- config_name: arb_Arab-swe_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-swe_Latn/train.parquet.gzip"
- config_name: dan_Latn-fil_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-fil_Latn/train.parquet.gzip"
- config_name: ell_Grek-pes_Arab
  data_files: 
  - split: train 
    path: "data/ell_Grek-pes_Arab/train.parquet.gzip"
- config_name: heb_Hebr-mya_Mymr
  data_files: 
  - split: train 
    path: "data/heb_Hebr-mya_Mymr/train.parquet.gzip"
- config_name: ell_Grek-zho_Hans
  data_files: 
  - split: train 
    path: "data/ell_Grek-zho_Hans/train.parquet.gzip"
- config_name: cat_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/cat_Latn-mya_Mymr/train.parquet.gzip"
- config_name: ces_Latn-hin_Deva
  data_files: 
  - split: train 
    path: "data/ces_Latn-hin_Deva/train.parquet.gzip"
- config_name: arb_Arab-plt_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-plt_Latn/train.parquet.gzip"
- config_name: ind_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-por_Latn/train.parquet.gzip"
- config_name: ces_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ces_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: kor_Hang-tur_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-tur_Latn/train.parquet.gzip"
- config_name: rus_Cyrl-zho_Hant
  data_files: 
  - split: train 
    path: "data/rus_Cyrl-zho_Hant/train.parquet.gzip"
- config_name: pan_Guru-zho_Hant
  data_files: 
  - split: train 
    path: "data/pan_Guru-zho_Hant/train.parquet.gzip"
- config_name: plt_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/plt_Latn-pan_Guru/train.parquet.gzip"
- config_name: mkd_Cyrl-pan_Guru
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-pan_Guru/train.parquet.gzip"
- config_name: mya_Mymr-nld_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-nld_Latn/train.parquet.gzip"
- config_name: hin_Deva-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/hin_Deva-rus_Cyrl/train.parquet.gzip"
- config_name: mya_Mymr-por_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-por_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-swe_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-swe_Latn/train.parquet.gzip"
- config_name: amh_Ethi-mya_Mymr
  data_files: 
  - split: train 
    path: "data/amh_Ethi-mya_Mymr/train.parquet.gzip"
- config_name: ben_Beng-ron_Latn
  data_files: 
  - split: train 
    path: "data/ben_Beng-ron_Latn/train.parquet.gzip"
- config_name: por_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-yor_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-dan_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-dan_Latn/train.parquet.gzip"
- config_name: pol_Latn-srp_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-srp_Latn/train.parquet.gzip"
- config_name: npi_Deva-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/npi_Deva-rus_Cyrl/train.parquet.gzip"
- config_name: dan_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-ita_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-kor_Hang
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-kor_Hang/train.parquet.gzip"
- config_name: pes_Arab-tur_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-tur_Latn/train.parquet.gzip"
- config_name: ayr_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-pol_Latn/train.parquet.gzip"
- config_name: dan_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/dan_Latn-por_Latn/train.parquet.gzip"
- config_name: ron_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/ron_Latn-swh_Latn/train.parquet.gzip"
- config_name: nld_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/nld_Latn-zho_Hans/train.parquet.gzip"
- config_name: arb_Arab-zho_Hans
  data_files: 
  - split: train 
    path: "data/arb_Arab-zho_Hans/train.parquet.gzip"
- config_name: deu_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/deu_Latn-nld_Latn/train.parquet.gzip"
- config_name: ces_Latn-ell_Grek
  data_files: 
  - split: train 
    path: "data/ces_Latn-ell_Grek/train.parquet.gzip"
- config_name: ayr_Latn-fil_Latn
  data_files: 
  - split: train 
    path: "data/ayr_Latn-fil_Latn/train.parquet.gzip"
- config_name: pol_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/pol_Latn-urd_Arab/train.parquet.gzip"
- config_name: eng_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/eng_Latn-mya_Mymr/train.parquet.gzip"
- config_name: arb_Arab-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/arb_Arab-mkd_Cyrl/train.parquet.gzip"
- config_name: eng_Latn-hin_Deva
  data_files: 
  - split: train 
    path: "data/eng_Latn-hin_Deva/train.parquet.gzip"
- config_name: ayr_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/ayr_Latn-urd_Arab/train.parquet.gzip"
- config_name: amh_Ethi-fra_Latn
  data_files: 
  - split: train 
    path: "data/amh_Ethi-fra_Latn/train.parquet.gzip"
- config_name: ind_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/ind_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: deu_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/deu_Latn-mya_Mymr/train.parquet.gzip"
- config_name: npi_Deva-srp_Latn
  data_files: 
  - split: train 
    path: "data/npi_Deva-srp_Latn/train.parquet.gzip"
- config_name: ita_Latn-npi_Deva
  data_files: 
  - split: train 
    path: "data/ita_Latn-npi_Deva/train.parquet.gzip"
- config_name: khm_Khmr-swh_Latn
  data_files: 
  - split: train 
    path: "data/khm_Khmr-swh_Latn/train.parquet.gzip"
- config_name: eng_Latn-ita_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-ita_Latn/train.parquet.gzip"
- config_name: cat_Latn-por_Latn
  data_files: 
  - split: train 
    path: "data/cat_Latn-por_Latn/train.parquet.gzip"
- config_name: ron_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/ron_Latn-zho_Hans/train.parquet.gzip"
- config_name: spa_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/spa_Latn-pan_Guru/train.parquet.gzip"
- config_name: por_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/por_Latn-sqi_Latn/train.parquet.gzip"
- config_name: kor_Hang-sqi_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-sqi_Latn/train.parquet.gzip"
- config_name: ell_Grek-nld_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-nld_Latn/train.parquet.gzip"
- config_name: fra_Latn-mos_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-mos_Latn/train.parquet.gzip"
- config_name: ita_Latn-ckb_Arab
  data_files: 
  - split: train 
    path: "data/ita_Latn-ckb_Arab/train.parquet.gzip"
- config_name: heb_Hebr-urd_Arab
  data_files: 
  - split: train 
    path: "data/heb_Hebr-urd_Arab/train.parquet.gzip"
- config_name: spa_Latn-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/spa_Latn-mkd_Cyrl/train.parquet.gzip"
- config_name: ben_Beng-ory_Orya
  data_files: 
  - split: train 
    path: "data/ben_Beng-ory_Orya/train.parquet.gzip"
- config_name: spa_Latn-hin_Deva
  data_files: 
  - split: train 
    path: "data/spa_Latn-hin_Deva/train.parquet.gzip"
- config_name: spa_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/spa_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: mkd_Cyrl-ory_Orya
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-ory_Orya/train.parquet.gzip"
- config_name: arb_Arab-tet_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-tet_Latn/train.parquet.gzip"
- config_name: ces_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/ces_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: arb_Arab-mya_Mymr
  data_files: 
  - split: train 
    path: "data/arb_Arab-mya_Mymr/train.parquet.gzip"
- config_name: npi_Deva-ron_Latn
  data_files: 
  - split: train 
    path: "data/npi_Deva-ron_Latn/train.parquet.gzip"
- config_name: eng_Latn-tet_Latn
  data_files: 
  - split: train 
    path: "data/eng_Latn-tet_Latn/train.parquet.gzip"
- config_name: spa_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-yor_Latn/train.parquet.gzip"
- config_name: ell_Grek-srp_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-srp_Latn/train.parquet.gzip"
- config_name: ces_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/ces_Latn-urd_Arab/train.parquet.gzip"
- config_name: swh_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/swh_Latn-urd_Arab/train.parquet.gzip"
- config_name: pes_Arab-pol_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-pol_Latn/train.parquet.gzip"
- config_name: spa_Latn-fra_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-fra_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-fil_Latn
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-fil_Latn/train.parquet.gzip"
- config_name: mkd_Cyrl-zho_Hant
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-zho_Hant/train.parquet.gzip"
- config_name: amh_Ethi-urd_Arab
  data_files: 
  - split: train 
    path: "data/amh_Ethi-urd_Arab/train.parquet.gzip"
- config_name: ita_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/ita_Latn-kor_Hang/train.parquet.gzip"
- config_name: mya_Mymr-tur_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-tur_Latn/train.parquet.gzip"
- config_name: ind_Latn-swh_Latn
  data_files: 
  - split: train 
    path: "data/ind_Latn-swh_Latn/train.parquet.gzip"
- config_name: fil_Latn-mya_Mymr
  data_files: 
  - split: train 
    path: "data/fil_Latn-mya_Mymr/train.parquet.gzip"
- config_name: ben_Beng-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/ben_Beng-jpn_Jpan/train.parquet.gzip"
- config_name: hun_Latn-rus_Cyrl
  data_files: 
  - split: train 
    path: "data/hun_Latn-rus_Cyrl/train.parquet.gzip"
- config_name: hun_Latn-nld_Latn
  data_files: 
  - split: train 
    path: "data/hun_Latn-nld_Latn/train.parquet.gzip"
- config_name: ell_Grek-eng_Latn
  data_files: 
  - split: train 
    path: "data/ell_Grek-eng_Latn/train.parquet.gzip"
- config_name: pol_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/pol_Latn-tur_Latn/train.parquet.gzip"
- config_name: ces_Latn-eng_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-eng_Latn/train.parquet.gzip"
- config_name: plt_Latn-ron_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-ron_Latn/train.parquet.gzip"
- config_name: ory_Orya-por_Latn
  data_files: 
  - split: train 
    path: "data/ory_Orya-por_Latn/train.parquet.gzip"
- config_name: plt_Latn-pol_Latn
  data_files: 
  - split: train 
    path: "data/plt_Latn-pol_Latn/train.parquet.gzip"
- config_name: ces_Latn-plt_Latn
  data_files: 
  - split: train 
    path: "data/ces_Latn-plt_Latn/train.parquet.gzip"
- config_name: bul_Cyrl-pes_Arab
  data_files: 
  - split: train 
    path: "data/bul_Cyrl-pes_Arab/train.parquet.gzip"
- config_name: kor_Hang-swe_Latn
  data_files: 
  - split: train 
    path: "data/kor_Hang-swe_Latn/train.parquet.gzip"
- config_name: amh_Ethi-mkd_Cyrl
  data_files: 
  - split: train 
    path: "data/amh_Ethi-mkd_Cyrl/train.parquet.gzip"
- config_name: npi_Deva-zho_Hant
  data_files: 
  - split: train 
    path: "data/npi_Deva-zho_Hant/train.parquet.gzip"
- config_name: pes_Arab-nld_Latn
  data_files: 
  - split: train 
    path: "data/pes_Arab-nld_Latn/train.parquet.gzip"
- config_name: mya_Mymr-srp_Latn
  data_files: 
  - split: train 
    path: "data/mya_Mymr-srp_Latn/train.parquet.gzip"
- config_name: arb_Arab-srp_Latn
  data_files: 
  - split: train 
    path: "data/arb_Arab-srp_Latn/train.parquet.gzip"
- config_name: arb_Arab-pan_Guru
  data_files: 
  - split: train 
    path: "data/arb_Arab-pan_Guru/train.parquet.gzip"
- config_name: fil_Latn-sqi_Latn
  data_files: 
  - split: train 
    path: "data/fil_Latn-sqi_Latn/train.parquet.gzip"
- config_name: swh_Latn-zho_Hans
  data_files: 
  - split: train 
    path: "data/swh_Latn-zho_Hans/train.parquet.gzip"
- config_name: fil_Latn-kor_Hang
  data_files: 
  - split: train 
    path: "data/fil_Latn-kor_Hang/train.parquet.gzip"
- config_name: fra_Latn-pan_Guru
  data_files: 
  - split: train 
    path: "data/fra_Latn-pan_Guru/train.parquet.gzip"
- config_name: mkd_Cyrl-swe_Latn
  data_files: 
  - split: train 
    path: "data/mkd_Cyrl-swe_Latn/train.parquet.gzip"
- config_name: fra_Latn-yor_Latn
  data_files: 
  - split: train 
    path: "data/fra_Latn-yor_Latn/train.parquet.gzip"
- config_name: hin_Deva-npi_Deva
  data_files: 
  - split: train 
    path: "data/hin_Deva-npi_Deva/train.parquet.gzip"
- config_name: spa_Latn-swe_Latn
  data_files: 
  - split: train 
    path: "data/spa_Latn-swe_Latn/train.parquet.gzip"
- config_name: ell_Grek-ckb_Arab
  data_files: 
  - split: train 
    path: "data/ell_Grek-ckb_Arab/train.parquet.gzip"
- config_name: swe_Latn-urd_Arab
  data_files: 
  - split: train 
    path: "data/swe_Latn-urd_Arab/train.parquet.gzip"
- config_name: sqi_Latn-tur_Latn
  data_files: 
  - split: train 
    path: "data/sqi_Latn-tur_Latn/train.parquet.gzip"
- config_name: ayr_Latn-jpn_Jpan
  data_files: 
  - split: train 
    path: "data/ayr_Latn-jpn_Jpan/train.parquet.gzip"
- config_name: ita_Latn-khm_Khmr
  data_files: 
  - split: train 
    path: "data/ita_Latn-khm_Khmr/train.parquet.gzip"
---
# Dataset Card for PolyNewsParallel

## Table of Contents
- [Table of Contents](#table-of-contents)
- [Dataset Description](#dataset-description)
  - [Dataset Summary](#dataset-summary)
  - [Uses](#uses)
  - [Languages](#languages)
- [Dataset Structure](#dataset-structure)
  - [Data Instances](#data-instances)
  - [Data Fields](#data-fields)
  - [Data Splits](#data-splits)
- [Dataset Creation](#dataset-creation)
  - [Curation Rationale](#curation-rationale)
  - [Source Data](#source-data)
  - [Data Collection and Processing](#data-collection-and-processing)
  - [Annotations](#annotations)
  - [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
  - [Social Impact of Dataset](#social-impact-of-dataset)
  - [Discussion of Biases](#discussion-of-biases)
  - [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
  - [Licensing Information](#licensing-information)
  - [Citation Information](#citation-information)

## Dataset Description

- **Homepage:** https://huggingface.co/datasets/aiana94/polynews-parallel
<!-- - **Repository:** https://github.com/andreeaiana/nase -->
<!-- - **Paper:**  -->
- **Point of Contact:** [Andreea Iana](https://andreeaiana.github.io/)
- **License:** [CC-BY-4.0-NC](https://creativecommons.org/licenses/by-nc/4.0/)

### Dataset Summary
PolyNewsParallel is a multilingual paralllel dataset containing news titles for 833 language pairs. It covers 65 languages and 17 scripts.

### Uses

This dataset can be used for domain adaptation of language models or machine translation.

### Languages

There are 65 languages avaiable:

|**Code**   |    **Language**             |   **Script**       |
|:-----------|:---------------------------|:-------------------|
| amh\_Ethi |   Amharic                 | Ethiopic          | 
| arb\_Arab |   Modern Standard Arabic  | Arabic            |
| ayr\_Latn |   Central Aymara          | Latin             |
| bam\_Latn |   Bambara                 | Latin             | 
| bbj\_Latn |   Ghomálá’                | Latin             | 
| ben\_Beng |   Bengali                 | Bengali           | 
| bul\_Cyrl |   Bulgarian               | Cyrillic          | 
| cat\_Latn |   Catalan                 | Latin             | 
| ces\_Latn |   Czech                   | Latin             | 
| ckb\_Arab |   Central Kurdish         | Arabic            | 
| dan\_Latn |   Danish                  | Latin             | 
| deu\_Latn |   German                  | Latin             | 
| ell\_Grek |   Greek                   | Greek             | 
| eng\_Latn |   English                 | Latin             | 
| est\_Latn |   Estonian                | Latin             | 
| ewe\_Latn |   Éwé                     | Latin             | 
| fil\_Latn |   Filipino                | Latin             |
| fin\_Latn |   Finnish                 | Latin             | 
| fon\_Latn |   Fon                     | Latin             | 
| fra\_Latn |   French                  | Latin             | 
| guj\_Gujr |   Gujarati                | Gujarati          | 
| hau\_Latn |   Hausa                   | Latin             | 
| heb\_Hebr |   Hebrew                  | Hebrew            |
| hin\_Deva |   Hindi                   | Devanagari        | 
| hun\_Latn |   Hungarian               | Latin             | 
| ibo\_Latn |   Igbo                    | Latin             | 
| ind\_Latn |   Indonesian              | Latin             | 
| ita\_Latn |   Italian                 | Latin             | 
| jpn\_Jpan |   Japanese                | Japanese          | 
| kaz\_Cyrl |   Kazakh                  | Cyrillic          | 
| khm\_Khmr |   Khmer                   | Khmer             |
| kor\_Hang |   Korean                  | Hangul            | 
| lav\_Latn |   Latvian                 | Latin             | 
| lit\_Latn |   Lithuanian              | Latin             | 
| lug\_Latn |   Ganda                   | Latin             | 
| luo\_Latn |   Luo                     | Latin             | 
| mkd\_Cyrl |   Macedonian              | Cyrillic          | 
| mos\_Latn |   Mossi                   | Latin             | 
| mya\_Mymr |   Burmese                 | Myanmar           | 
| nld\_Latn |   Dutch                   | Latin             | 
| npi\_Deva |   Nepali                  | Devanagari        | 
| ory\_Orya |   Odia                    | Oriya             | 
| pan\_Guru |   Eastern Panjabi         | Gurmukhi          | 
| pcm\_Latn |   Nigerian Pidgin         | Latin             | 
| pes\_Arab |   Western Persian         | Arabic            | 
| plt\_Latn |   Malagasy                | Latin             | 
| pol\_Latn |   Polish                  | Latin             | 
| por\_Latn |   Portuguese              | Latin             |
| ron\_Latn |   Romanian                | Latin             | 
| rus\_Cyrl |   Russian                 | Cyrillic          | 
| spa\_Latn |   Spanish                 | Latin             | 
| sqi\_Latn |   Albanian                | Latin             | 
| srp\_Latn |   Serbian                 | Latin             | 
| swe\_Latn |   Swedish                 | Latin             | 
| swh\_Latn |   Swahili                 | Latin             | 
| tet\_Latn |   Tetun                   | Latin             | 
| tsn\_Latn |   Tswana                  | Latin             | 
| tur\_Latn |   Turkish                 | Latin             | 
| twi\_Latn |   Twi                     | Latin             | 
| urd\_Arab |   Urdu                    | Arabic            | 
| wol\_Latn |   Wolof                   | Latin             |
| yor\_Latn |   Yorùbá                  | Latin             | 
| zho\_Hans |   Chinese                 | Han (Simplified)  | 
| zho\_Hant |   Chinese                 | Han (Traditional) | 
| zul\_Latn |   Zulu                    | Latin             | 

The heatmap shows the language pairs available, as well as the number of articles per language pair.

<center>
    <img src="https://huggingface.co/datasets/aiana94/polynews-parallel/resolve/main/polynews_parallel.png" alt="PolyNewsParallel: Number of texts per language pair">
</center>

## Dataset Structure

### Data Instances
```
>>> from datasets import load_dataset
>>> data = load_dataset('aiana94/polynews-parallel', 'eng_Latn-ron_Latn')

# Please, specify the language code,

# A data point example is below:

{
"src": "They continue to support the view that this decision will have a lasting negative impact on the rule of law in the country. ",
"tgt": "Ei continuă să creadă că această decizie va avea efecte negative pe termen lung asupra statului de drept în țară. ",
"provenance": "globalvoices"
}

```

### Data Fields

- src (string): source news text
- tgt (string): target news text
- provenance (string) : source dataset for the news example

### Data Splits

For all languages, there is only the `train` split.


## Dataset Creation

### Curation Rationale

Multiple multilingual, human-translated, datasets containing news texts have been released in recent years. 
However, these datasets are stored in different formats and various websites, and many contain numerous near duplicates. 
With PolyNewsParallel, we aim to provide an easily-accessible, unified and deduplicated parallel dataset that combines these disparate data sources, and which can be used for domain adaptation of language models or machine translation in both high-resource and low-resource languages.

### Source Data

The source data consists of five multilingual news datasets.

- [GlobalVoices](https://opus.nlpl.eu/GlobalVoices/corpus/version/GlobalVoices) (v2018q4)
- [WMT-News](https://opus.nlpl.eu/WMT-News/corpus/version/WMT-News) (v2019)
- [MAFAND](https://huggingface.co/datasets/masakhane/mafand) (`train` split)

#### Data Collection and Processing

We processed the data using a **working script** which covers the entire processing pipeline. It can be found [here](https://github.com/andreeaiana/nase/script/polynews).

The data processing pipeline consists of:
1. Downloading the WMT-News and GlobalVoices News from OPUS.
2. Loading MAFAND datasets from Hugging Face Hub (only the `train` splits).
4. Concatenating, per language, all news texts from the source datasets.
5. Data cleaning (e.g., removal of exact duplicates, short texts, texts in other scripts)
6. [MinHash near-deduplication](https://github.com/bigcode-project/bigcode-dataset/blob/main/near_deduplication/minhash_deduplication.py) per language.


### Annotations 

We augment the original samples with the `provenance` annotation which specifies the original data source from which a particular examples stems.


#### Personal and Sensitive Information

The data is sourced from newspaper sources and contains mentions of public figures and individuals.


## Considerations for Using the Data

### Social Impact of Dataset
[More Information Needed]


### Discussion of Biases
[More Information Needed]


### Other Known Limitations

Users should keep in mind that the dataset contains short news texts (e.g., mostly titles), which might limit the applicability of the developed systems to other domains.


## Additional Information 

### Licensing Information
The dataset is released under the [CC BY-NC Attribution-NonCommercial 4.0 International license](https://creativecommons.org/licenses/by-nc/4.0/).

### Citation Infomation

**BibTeX:**

[More Information Needed]