Other Model Architectures

[michaelholm-ce]

Hello -- Will you please share guidance on how to apply the quantization on other model architectures, such as Squeezenet? Thank you in advance.

[jiaxiang-wu]

@michaelholm-ce You need to implement your own training/quantization process to convert an arbitrary pre-trained network to its quantized counterpart. The mathematics details can be found in our CVPR '16 paper. For now, I could not publish this part of code due to some issues. Sorry for that.