Benedikt Fuchs

The first part is almost finished: https://github.com/flairNLP/flair/pull/2643 is ready for review. That one surprisingly straight forward: First think of how to refactor a model and then apply the same to...

As long as you are not using the flair embeddings with flair version < 0.11, you can apply dynamic quantisation on all flair models that run on cpu. However, you...

*ping* to reveive issue as it isn't dead

Hi @edoust , sorry for the late answer I think it will take a long time to finish this. So far the models can be exported without embeddings and the...

The predictions are on token level. You can choose how the embedding for a token is aggregated if it consists of several subtokens. This is done by setting the `subtoken_pooling`...

Hello, judging from your loss, I would assume that your hyperparameters are not a good choice for your dataset. Usually the lerning rate is the most important parameter to tune....

Hey @Newlearner012 For dates I would consider using the [regex tagger](https://github.com/flairNLP/flair/blob/master/flair/models/regexp_tagger.py#L44) crafting a few regexes should be usefull to cover 99% of cases

Hello, I made a PR which should make it possible to use `torch.jit.trace` and therefore also aws neuron, can you take a look at https://github.com/flairNLP/flair/pull/2739/files#diff-12961c27a5afbdccd88c22a4678b771edbe9a64e512fd5db7b3ac4df4fa4234b and see if that works?

Hello, I think this is due to a typo in [this line](https://github.com/flairNLP/flair/blob/master/flair/embeddings/base.py#L435) which should be `if "operation" in state:`

> Am I missing something? I think you are missing, that Flair creates Flair-models and not hugging face models. You can see how to use them by clicking on the...