Bias initialization

[JohnMBrandt]

The original focal loss paper initialize the bias on the final sigmoid conv layer like so:

init = tf.constant_initializer([-np.log(0.99/0.01)])

so that negative examples will start training with no loss, and positive examples will have very high loss. Did you experiment with this bias setting? I could not find good results with it with FTL.

[nabsabraham]

Hi John, I initialize all weights to follow glorot_normal distribution but did not experiment with the original paper's init method. Is there a large difference in DSC when using the init versus not using it?