Open mustafaIhssan opened 6 years ago
Yeah, I might upload it as a kaggle dataset when I get time...
On Tue, Jun 5, 2018, 17:51 Mustafa Ihssan A. Naji notifications@github.com wrote:
The project is very cool, I have an idea to improve bnn v3 ( RGB -> Bee count ) and i would like to try it.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/matpalm/bnn/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AABULdvCMc2dTvHdrkojiIivYG5lExGCks5t5jiZgaJpZM4UaUch .
Great, Thanks alot
I look forward to the dataset aswell, to learn from this project
Hi, I just started gathering a dataset and have uploaded it to kaggle: https://www.kaggle.com/jonathanbyrne/to-bee-or-not-to-bee
There are two DB files with the first 100 or so images labelled. The images may be too close as I was wondering if it is possible to spot varroa with a picam. I scaled the images to a quarter of the size to speed up the labelling but if you want the full size images just let me know!
The project is very cool, I have an idea to improve bnn v3 ( RGB -> Bee count ) and i would like to try it. The Idea is using the trained v2 as base of v3, roughly i'm thinking of removing the softmax and argmax of v2 and adding classification layer, and maybe some skip connection from the bottleneck.