adaptive

Hi Jiahui,

Thanks for the great work. I'm trying to reproduce AutoSlim for CIFAR-10 (Table 2).
Could you please provide a detailed hyperparameter you used for it?

I'm able to train the baseline MobileNetV2 1.0x to 7.9 Top-1 error using the following hyperparameters:

• 0.1 initial learning rate
• linear learning rate decay
• 128 batch size
• 300 epochs of training
• 5e-4 weight decay

When computing the isosurface, when 4 simplex edges crosses the isosurface, we add 2 triangles to the isosurface triangulation.

Right now we add these arbitrarily (in these 3 lines) but we should add them so that the curvature is minimized.

The current state o