romilly
commented
7 years ago 1) That paper only used 5-10 categories, but that was not due to limitations of the CNN; it was a choice made by the researchers reflecting the nature of the data that they used. 2) That paper has been superceded by https://arxiv.org/abs/1509.01626 3) I do not know of any high limit. The approach is constrained by available data and the computational power requited.
On 9 March 2017 at 08:28, 郭同jet notifications@github.com wrote:
We can use CNN to classify more than 10 thousands of images of the ImageNet. I find that CNN could only classify 10-20 text classes as this paper https://arxiv.org/pdf/1502.01710.pdf write. So what is the high limit of short text classification's classes' number?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/cs231n/cs231n.github.io/issues/130, or mute the thread https://github.com/notifications/unsubscribe-auth/AAJHPt6giY0xIwOj0U0coH3ONq8TgB5mks5rj7gQgaJpZM4MXx2P .
-- personal:@romillyc work:@rareblog skype:romilly.cocking web: http://blog.rareschool.com/
We can use CNN to classify more than 10 thousands of images of the ImageNet. I find that CNN could only classify 10-20 text classes as this paper write. So what is the high limit of short text classification's classes' number?