Sathyaprakash Narayanan

The link seems to be still invalid.

Found the paper:https://dl.acm.org/doi/abs/10.1145/3007787.3001163. Also it has been shown in various other papers and has also been tested here: [https://github.com/satabios/sconce/blob/main/tutorials/Pruning.ipynb](https://github.com/satabios/sconce/blob/main/tutorials/Pruning.ipynb)

@VainF can you comment on why the pruning fails on mobilenets?

I followed [Saving and loading models across devices in PyTorch](https://pytorch.org/tutorials/recipes/recipes/save_load_across_devices.html). The issue isn't during saving or reloading. However, it is with porting from one device to another, as mentioned above,...

Either way, the bug persists. When the model is in memory and is prompted to transfer to a different device (say from GPU to CPU or vice versa) or when...

Yes, Pruning is an invasive process. However, if you can find the sweet spot (i.e...) the tradeoff between model degradation and removing redundant data. We can cram this space, to...

To add to the above point the final result table gives a glimpse of the technique quantitatively. Also note that there is actual reduction of MAC operations unlike quantization. 

That would be great.

@ffdm, Thanks for checking this tutorial. The sconce package was updated post to the the PR, hence if you rerun the tutorial you would find Quantization also being applied on...