Enrique Manjavacas

From the last discussion,t we concluded to leave out the stuff coming from 3D party libraries (including the delayed stuff)

Could you pin point what is it that the Delayer class is achieving? I see a lot of new code involved in printing stuff and so on. Perhaps there is...

Ye, go ahead. Only I'd rather really keep it simple. For instance if there is that class that delays the step, we don't need to add a bunch of new...

Check the last commit, that might solve the issues

Any chance you could quickly replicate the results for the combination that was on par with ReduceLROnPlateau but was faster? (No rush here) On Mon, Dec 7, 2020 at 6:25...

Hey both, Indeed, I don't have the capacities to keep pie updated to the last pytorch versions. However, I just pip installed it (`pip install nlp-pie`) and ran the tests...

Hey, here you have an example config from a trained model. You should use transformer.py, instead of train.py. [transformer-config.json.gz](https://github.com/emanjavacas/pie/files/5149637/transformer-config.json.gz)

Hi both, We could just increase the default max_seq_len to accommodate. However, this decoder is only meant for tasks like lemmatization. It's hard to imagine a language with lemmas longer...

Yup, don't use the "attentional" decoder for POS, use "linear" as shown in the example config for PoS https://github.com/emanjavacas/pie/blob/master/configs/pos.json#L12 On Fri, Oct 27, 2023 at 12:02 PM Vincent Prins ***@***.***>...

Hey, Thanks for the report. There is some code that makes sure to store the version of pie that are installing during setup. We record the commit version of the...