Jonna Matthiesen

icon company Embedl AB icon location Deep Learning Researcher @ Embedl | Hardware-Aware Model Optimization - NAS - Pruning - Quantization

Results 2 comments of Jonna Matthiesen

About the number of epochs?

Regarding the total number of epochs and the LR schedule the paper states that _"We start with a learning rate of 0.1, divide it by 10 at 32k and 48k...

update mean and std of cifar dataset

FYI: The previous values of mean and std correspond to the ImageNet normalization values.