Closed prabhatM closed 4 years ago
Hello @prabhatM interesting idea - our embedding stacks could be used in other sequence tagging architectures next to BiLSTM, so this would be possible. From our side, we probably won't implement ID-CNN in the near future, but if you would like to add this to Flair, we'd greatly appreciate it!
Hi Allan,
Thank you for creating an interesting framework - Flair. I am just getting hang of it. I am trying to create a custom model for clinical text. In future, I would try IDCNN and update you.
Happy New Year!
Regards
Prabhat MOHANTY
On Mon, 31 Dec 2018, 16:43 Alan Akbik <notifications@github.com wrote:
Hello @prabhatM https://github.com/prabhatM interesting idea - our embedding stacks could be used in other sequence tagging architectures next to BiLSTM, so this would be possible. From our side, we probably won't implement ID-CNN in the near future, but if you would like to add this to Flair, we'd greatly appreciate it!
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/zalandoresearch/flair/issues/342#issuecomment-450634036, or mute the thread https://github.com/notifications/unsubscribe-auth/AG6jXHB4XPvzeg93a-1600VD4aOnA4cmks5u-fFOgaJpZM4ZlAjP .
Great, thanks - I look forward to hearing about your experience!
Happy new year!
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.
Since Flair has taken the stacked approach, will there be any possibility of support for Iterated dilated CNN architecture (ID-CNN) in near future?
Regards
Prabhat MOAHNTY