Pos tagging nlp1/12/2024 ![]() Your suggestion didn’t work when I tried editing the source recipe file ie. ValueError: tuple.index(x): x not in tuple Sorry I did not give you the concrete error during pos.batch-train:įile “cython_src/prodigy/models/pos.pyx”, line 90, in .batch_trainįile “cython_src/prodigy/models/pos.pyx”, line 136, in .updateįile “cython_src/prodigy/models/pos.pyx”, line 144, in .inc_gradientįile “cython_src/prodigy/models/pos.pyx”, line 79, in .get_label_index Please let me know if this is possible currently or will be soon. Ideally, we would like to train a tagger with arbitrary tags. I realize I can use NER for this case, but that's a more complicated model. ValueError: Resizing pre-trained Tagger models is not currently supported. I get this error even though there was no pre-trained model: I noticed that even with en_vectors_web_lg, the default model has 57 output tags. BTW, it would be very helpful if you could document the input and output of the Cythonized methods. I inspected the code as much as is possible (calling help() and interacting with it. The dictionary must have an entry POS whose value is one of the Universal Dependencies tags. The keys of the tag map should be strings in your tag set. The spacy documentation says that the keys should be tags from our set which they are: I tried making a custom tag map as follows: For more specific use cases, you can now even pass in a JSON tag map that maps coarse-grained to fine-grained tags. Sorry again! The pos.teach and pos.batch-train recipes should now also be able to tell whether you want to use fine-grained or coarse-grained labels (by comparing the labels, so you don’t have to tell it explicitly). Just released v1.5.1, which should fix the underlying problems. Errors with pos.teach and pos.batch-train.
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |