Datasets:

Helsinki-NLP
/

tatoeba_mt

Converts this dataset from a Python loading script + raw .tsv files to data-only Parquet (the datasets-cli convert_to_parquet command that used to automate this has been removed from the datasets library, and script-loading support itself was dropped in v4, so this was built by hand: ran the maintainer's own tatoeba_mt.py once under datasets<4.0 to produce the Parquet files, then assembled this PR).

Removes the .py loading script (archived on the script branch)
Adds a configs:/dataset_info block to the README so the dataset loads with plain datasets>=4 and no trust_remote_code
Enables the dataset viewer

Configs included in this commit: 783.

Convert dataset to Parquet (drop loading script)84b73b94

Convert dataset to Parquet (drop loading script) (part 2)1ed29625

Convert dataset to Parquet (drop loading script) (part 3)a3560442

Convert dataset to Parquet (drop loading script) (part 4)bbad168d

Convert dataset to Parquet (drop loading script) (part 5)d7e0cda3

shenbinqian changed pull request status to open Sep 20

tiedeman

Helsinki-NLP Research Group org Sep 21

Merging this now but there will be a new release and update coming soon in this dataset.

tiedeman changed pull request status to merged Sep 21

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment