I installed the new update 1.5.1. on my windows 10, 64bit system in an anaconda virtual environment.
Installation went fine.
I then set up the de_core_web_sm and added a new dataset by “python -m prodigy dataset ensi_set “ENSI-2017” --author RalfBendzko”.
Then I tried to “python -m prodigy textcat.teach ensi_set de_core_news_sm G:\test.txt” but got:
ValueError: Buffer and memoryview are not contiguous in the same dimension.
How to proceed?
I just tested it and I think I know what the problem is: When you run
textcat.teach, you’re not specifying a
--label you want to annotate. This seems to throw off the text classifier and causes this error. Prodigy should probably handle this better and warn or raise a more descriptive error if no labels are set. The following should work as expected:
python -m prodigy textcat.teach ensi_set de_core_news_sm G:\test.txt --label SOMETHING
If you haven’t done so already, you probably also want to create some match patterns to help you pre-select examples, especially if you’re training a new category from scratch. (You can find more details on this in your
Thanks a lot for the fast response.
You are right, with the label added everything just worked fine:)
15 posts were merged into an existing topic: prodigy.json setting causes web app error