multi30k

Multi30k This dataset contains the “multi30k” dataset, which is the “task 1” dataset from here. Each example consists of an “en” and a “de” feature. “en” is an English sentence, and “de” is the German translation of the English sentence. Data Splits The Multi30k dataset has 3 splits: train, validation, and test. Dataset Split Number of Instances in Split Train 29,000 Validation 1,014 Test 1,000 Citation Information… See the full description on the dataset page:

Source: Hugging Face Hub (bentrevett/multi30k). Metadata imported from the dataset’s Hub tags.