Use new Hugging Face datasets library

[stefan-it]

Hi,

Hugging Face released a new awesome library: datasets. It contains "Datasets and evaluation metrics for natural language processing":

https://github.com/huggingface/datasets

I'm currently working on integrating some NER datasets.

So the main aim would be to use the datasets library instead of writing own dataset classes in Flair :)

[stefan-it]

I will start on integrating the GermEval 2014 dataset into nlp.

Then the next step will be to use this dataset from nlp instead of our own:

flair/flair/datasets/sequence_labeling.py

Line 579 in f41ee04

class GERMEVAL_14(ColumnCorpus):

class.

[alanakbik]

Awesome!

[stefan-it]

GermEval 14 is now available (latest master version of nlp) - WNUT 17 is on its way. So I'll think about a nice integration of it into Flair.

I'm not sure if it's ready for the upcoming 0.5 release of Flair, because there are discussions about the name nlp, see this issue 😅

[alanakbik]

Oh yes makes sense to change the name to something more specific ;)

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.