Datasets:
Tasks:
Translation
Size:
1K<n<10K
Dataset Viewer
The dataset viewer is not available for this subset.
The default config contains 104 while it should generally contain 3 splits maximum (train/validation/test). If the splits am_om, am_ti, ar_es_LA, ar_fr, ar_hi... are not used to differentiate between training and evaluation, please consider defining configs of this dataset instead. You can find how to define configs instead of splits here: https://huggingface.co/docs/hub/datasets-data-files-configuration
Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
TICO-19 Dataset
The original dataset is available at tico-19.github.io.
Available language pairs:
- am_om
- am_ti
- ar_es_LA
- ar_fr
- ar_hi
- ar_id
- ar_pt_BR
- ar_ru
- ar_zh
- en_ar
- en_bn
- en_ckb
- en_din
- en_es_LA
- en_fa
- en_fr
- en_fuv
- en_ha
- en_hi
- en_id
- en_km
- en_kr
- en_ku
- en_lg
- en_ln
- en_mr
- en_ms
- en_my
- en_ne
- en_nus
- en_om
- en_prs
- en_ps
- en_pt_BR
- en_ru
- en_rw
- en_so
- en_sw
- en_ta
- en_ti_ER
- en_ti_ET
- en_tl
- en_ur
- en_zh
- en_zu
- es_LA_ar
- es_LA_fr
- es_LA_hi
- es_LA_id
- es_LA_pt_BR
- es_LA_ru
- es_LA_zh
- fa_prs
- fr_ar
- fr_es_LA
- fr_hi
- fr_id
- fr_lg
- fr_ln
- fr_pt_BR
- fr_ru
- fr_rw
- fr_sw
- fr_zh
- fr_zu
- hi_ar
- hi_bn
- hi_es_LA
- hi_fr
- hi_id
- hi_mr
- hi_pt_BR
- hi_ru
- hi_ur
- hi_zh
- id_ar
- id_es_LA
- id_fr
- id_hi
- id_pt_BR
- id_ru
- id_zh
- ku_ckb
- pt_BR_ar
- pt_BR_es_LA
- pt_BR_fr
- pt_BR_hi
- pt_BR_id
- pt_BR_ru
- pt_BR_zh
- ru_ar
- ru_es_LA
- ru_fr
- ru_hi
- ru_id
- ru_pt_BR
- ru_zh
- zh_ar
- zh_es_LA
- zh_fr
- zh_hi
- zh_id
- zh_pt_BR
- zh_ru
Usage example
from datasets import get_dataset_split_names, load_dataset
# Get list of splits without loading the data
splits = get_dataset_split_names("atepeq/tico19")
print("\nAvailable splits:")
print(splits)
# Find first dataset containing 'en' in its name
en_dataset = next((split for split in splits if 'en' in split), None)
if en_dataset:
print(f"\nLoading {en_dataset} split...")
# Load the entire dataset but only access the split we need
dataset = load_dataset("atepeq/tico19", split=en_dataset)
print(f"\nFirst 10 records from {en_dataset} split:")
for i, record in enumerate(dataset.select(range(10))):
print(f"\n{i+1}. Source: {record['source']}")
print(f" Translation: {record['translation']}")
else:
print("\nNo dataset found containing 'en' in its name")
- Downloads last month
- 82