Some confusions about code

[zongjuede]

Thanks for your paper and sharing code. I have some confusions about code, can you explain it for me.

1. In function channel_selection()， why the original model is replaced by pruned model? Is it since the loaded pre-trained model without fine-tuning.
2. In function update_model, why the network is splited repeatly(the network will be split a time if the update_model() be called)？But the segment seems same after each split,
3. Is it original net just be a baselone. Why origin net is continualy updated(line 503 in channel_pruning.py) if it is just be a baseline ?
   Thank you very much.
   Expect for yor reply.

[liujingcs]

1. To reduce the discrepancy between the original model and pruned model, we replace the original model with the fine-tuned model at the first stage.
2. After channel selection, the pruned model has changed. Therefore, we need to update the pruned model in the trainer.
3. Except in the first stage of fine-tuning, we fix the original model during channel pruning.