Ananiya
07/26/2022, 3:12 PMRobertaTokenizerFast
in pytho..
In kotlin
I was managed to get the size and some information using `OnnxInferenceModel`(kotlindl) but I am quite confused on vectorizing the sentence because i don't have access to the vocabulary, the approach used the vectorize the sent.. or the dimensions to implement the vectorizer manualy
So I was wondering if there's any possible way or library that could workaroundIlya Muradyan
07/26/2022, 4:30 PMzaleslaw
07/27/2022, 12:06 PMAnaniya
07/27/2022, 2:23 PMroman.belov
07/29/2022, 10:26 AMAnaniya
07/29/2022, 4:44 PMHampus Londögård
08/01/2022, 5:50 AMAnaniya
08/01/2022, 5:52 AMHampus Londögård
08/29/2022, 8:01 AMClassifierPipeline
& TokenClassificationPipeline
which means that Text. Classiification and Token Classiification (e.g. NER) is supported now 👍
Files can be loaded from file-system (ONNX & PyTorch) and from HuggingFace Hub (ONNX)Ananiya
08/29/2022, 8:20 AM