sigver
sigver copied to clipboard
Transfer learning
Is transfer learning with using the pretrained signet model a good idea? Adding additional data which is specific to current application and updating the cnn model?
For other similar tasks, yes, this makes sense.