CoolFace
Datasetpublic

Ilia-Iliev/romani-sketches-dpo

This dataset consists of parallel text pairs for translating sentences from colloquial spoken Romani to Bulgarian. The style is conversational and the samples are taken from short dialogue videos. It is formatted as instruction-completion pairs suitable for training machine translation models.

sourceHugging Faceupdated 5mo agoView on Hugging Face
0likes11downloads
7 commits on main
96a8e195mo ago

Upload data/romani_prompts_README.md with huggingface_hub

Ilia-Iliev
eb27f8d5mo ago

Create README.md

Ilia-Iliev
22066125mo ago

Upload comprehension_dpo.jsonl with huggingface_hub

Ilia-Iliev
0d0ed2a5mo ago

Create README.md

Ilia-Iliev
9fd9d955mo ago

Upload comprehension_dpo.jsonl with huggingface_hub

Ilia-Iliev
03d6d8d5mo ago

Upload comprehension_dpo.jsonl with huggingface_hub

Ilia-Iliev
c12dd645mo ago

initial commit

Ilia-Iliev