Emil Hvitfeldt comments

Repositories
Issues
Comments

Results 253 comments of


                                            Emil Hvitfeldt

trafficstars

Using pretrained models in step_tokenize_sentencepiece

first part: the reason why you are getting characters when you were expecting BPE is because the vocabulary size is set to small. With a `vocabulary_size` of 25 it is...

Using pretrained models in step_tokenize_sentencepiece

it is not currently possible to use a pretrained model, but it is a valid thing to want to do.

pass calls through `forge_predictors()`

without spending too much time right now, i suppect it might be somewhere in {hardhat} that is the issue. Here is the traceback ```r ▆ 1. ├─stats::predict(wf_fit, mtcars %>% select(-disp))...