Loading DenseNet Checkpoints for CIFAR-10/100

[superctj]

Thank you for open sourcing this great project. The way you saved the Cifar10/100 checkpoints makes it hard to reuse the pre-trained Densenets in another project. Could you please release the checkpoints that can be loaded like

deep_Mahalanobis_detector/OOD_Baseline_and_ODIN.py

Line 49 in 90c2105

model = models.DenseNet3(100, int(args.num_classes))

[superctj]

Never mind. The workaround is to keep the same directory structure and load the entire model

[saandeepa93]

Can you please let me know the solution for this?