Datasets:
Tasks:
Feature Extraction
Modalities:
Text
Formats:
parquet
Languages:
French
Size:
10K - 100K
License:
metadata
language:
- fr
license: apache-2.0
size_categories:
- 10K<n<100K
task_categories:
- feature-extraction
pretty_name: ordalie-fr-sts-benchmark
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: score
dtype: int64
splits:
- name: test
num_bytes: 14934570
num_examples: 10000
download_size: 9328832
dataset_size: 14934570
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
Ordalie - French STS Benchmark
- 30k sentence pairs
- Score either 0 or 1