Licensing question: SpaCy pipelines vs training datasets #13579
surdina
started this conversation in
Language Support
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
The German SpaCy pipelines are listed under an MIT license. However, sources appear to include the TIGER corpus, with a license that requires the licensee to be part of an academic or educational institution.
Does this mean that in practice, the MIT license is void because one of the data sets is released under a stricter license? Or does the pre-trained pipeline not automatically inherit any training set's license restrictions?
Beta Was this translation helpful? Give feedback.
All reactions