How to train your own Object Detector with TensorFlow’s Object Detector API
Dat Tran
3.3K125

Hi Dat,

I have a basic query on the process involved during training.

Are we,

  1. re-training only the classifier in the final layer on the new image dataset (or)
  2. fine-tuning the weights for all the layers, both the dense and sparse ones ?
One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.