• Post Reply Bookmark Topic Watch Topic
  • New Topic
programming forums Java Mobile Certification Databases Caching Books Engineering Micro Controllers OS Languages Paradigms IDEs Build Tools Frameworks Application Servers Open Source This Site Careers Other all forums
this forum made possible by our volunteer staff, including ...
Marshals:
  • Campbell Ritchie
  • Paul Clapham
  • Ron McLeod
  • Liutauras Vilda
  • Bear Bibeault
Sheriffs:
  • Jeanne Boyarsky
  • Tim Cooke
  • Devaka Cooray
Saloon Keepers:
  • Tim Moores
  • Tim Holloway
  • Piet Souris
  • salvin francis
  • Stephan van Hulst
Bartenders:
  • Frits Walraven
  • Carey Brown
  • Jj Roberts

TensorFlow 2.0 in Action: Max Number of Labels

 
Ranch Hand
Posts: 35
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Hi Thushan.

What is the practical maximum limit for the number of labels a multi-label project can work with?
Are we talking tens, hundreds, thousands, more?

This question applies to all types of input data, but I'm thinking specifically of Natural Language Processing, so Bert may have some limits too. Or even XBert.


Thanks
Don.
 
Author
Posts: 24
2
  • Likes 1
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Hi Don,

Interesting question. I don't think theoretically there's a limit. However, practically I believe so.

1. Having too many labels might result in training instabilities
2. If the maximum value of the label is large, you might run out of memory creating that layer

Hope this helps
 
Don't get me started about those stupid light bulbs.
reply
    Bookmark Topic Watch Topic
  • New Topic