Keras-RetinaNet-for-Open-Images-Challenge-2018
Keras-RetinaNet-for-Open-Images-Challenge-2018 copied to clipboard
Inferring on Level 2 by the model
The repo is easy to use but there was one thing I wanted to check. I wanted to make inference on Level 2 and not Level 1. Hence I was thinking is it possible to do, this without having to train a model by keeping your model as backbone.
You're going to have to train the model for Level 2. I tried giving it images from labels of Level 2 and it won't predict it. Also, as the name suggests the model trained on Level 1 only.