Datasets:
Tasks:
Sentence Similarity
Modalities:
Text
Formats:
json
Sub-tasks:
semantic-similarity-scoring
Size:
10K - 100K
ArXiv:
License:
Added utf-8 encoding to gszip open as STS17 has multilingual data
#2
by
rajathravi - opened
The data generator fails when we try to evaluate the STS17 benchmark across all languages. Hence, added utf-8 encoding to gszip open as STS17 has multilingual data.
Hello!
Ah, yes. This can indeed cause issues on Windows devices. Thanks for the fix!
- Tom Aarsen
tomaarsen changed pull request status to
merged