What is the proper way of using last_epoch in a lr_scheduler?

[Shisho_Sama (A curious guy here!)]

Asked the new question concerning the out of memory issue in 0.4 :

Training fails by out of memory error on Pytorch 0.4 but runs fine on 0.3.1

hi, I recently updated my Pytorch installation to the latest version which is 0.4. I installed the newer version like this : me@shishosama:/media/me/tmpstore/SimpNet_PyTorch$ pip install http://download.pytorch.org/whl/cu90/torch-0.4.0-cp36-cp36m-linux_x86_64.whl Collecting torch==0.4.0 from http://download.pytorch.org/whl/cu90/torch-0.4.0-cp36-cp36m-linux_x86_64.whl Downloading http://download.pytorch.org/whl/cu90/torch-0.4.0-cp36-cp36m-linux_x86_64.whl (566.4MB) 100% |█████████████████…