Can this dataset be used commercially through another third party library?

[samarth12]

Hi,

I am interested in using some language models built using this particular dataset in Spanish. These language models were trained by Spacy and Stanza.

Here is the link to the model made available by Spacy. Spacy itself as a library is available to use commercially MIT, which means I can use their Spanish language model but the dataset it uses is GPL.

Here is the link to the model made available by Stanza (scroll down to Spanish). Stanza is available for commercial use through Apache 2.0 as well.

I am interested to know if the Spanish AnCora is available through a commercial license? Or if the dataset can be used commercially if it used to train language models by another company/library that offers its services commercially.

I would appreciate any help or direction, thank you!

[dan-zeman]

Any licensing questions have to be directed to the providers of the original pre-UD dataset at the University of Barcelona. I'm afraid that they may not be monitoring this issue tracker. The website of the AnCora corpus is here.