| |
This archive contains 36 data sets derived from the World Atlas of Language Structures (WALS), formatted for use with RoBERTa-based models in linguistic typology and cross-linguistic prediction tasks.
Since I don’t have access to the actual contents of that ZIP file, I’ll assume it contains processed into a format compatible with RoBERTa (e.g., preprocessed feature sets or training splits for linguistic typology tasks). WALS Roberta Sets 1-36.zip
Here’s a you could include as a README.txt inside the zip: README: WALS Roberta Sets 1-36 This archive contains 36 data sets derived from