Beyond_Deep_Ensembles

Beyond_Deep_Ensembles copied to clipboard

Training loss does not converge after reloading checkpoint to continue training

1

Hi, I'm trying to reload my pre-trained network to continue training in Cifar-10 experiments `cifar.py`, while the loss does not converge after reloading (the loss still converges if the model...

YitianShi

Missing/wrong hparams in cifar.yaml

1

Hello, Thanks for the great work. When trying to replicate your experiments I noticed the "update_interval" parameter is missing for CIFAR10's swag optimizer. There are some other minor typos like...

emirkonuk