nlj-guang / quantize_cnn Goto Github PK

Hello
I wanted to test quantization, i was so appreciated to find your code
But, As i couldn`t read chineese, It was a little difficult to understand meaning of parameters
Could you tell me how to run this code?

I got this error, and how i can handle this?
and actually i don`t know what is the role of the line

path = './cifar_net.pth' //i change h.path to path
act_path = os.path.join(path, 'temp', name+'_'+data_set+'_act', name+'_act.pth.tar')
init_model = torch.load(act_path)
model.load_state_dict(init_model['state_dict'])

$ python3 train.py --save-dir ./result

=> init_lr 0.001
=> lr-step 6
=> momentum 0.9
=> weight-decay 0.0001
=> batch-size 128
=> balance 100
=> save-dir ./result

Traceback (most recent call last):
File "train.py", line 151, in
main()
File "train.py", line 86, in main
init_model = torch.load(act_path)
File "/home/sr2/so0.bak/venv/venv/lib/python3.7/site-packages/torch/serialization.py", line 584, in load
with _open_file_like(f, 'rb') as opened_file:
File "/home/sr2/so0.bak/venv/venv/lib/python3.7/site-packages/torch/serialization.py", line 234, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/home/sr2/so0.bak/venv/venv/lib/python3.7/site-packages/torch/serialization.py", line 215, in init
super(_open_file, self).init(open(name, mode))
NotADirectoryError: [Errno 20] Not a directory: './cifar_net.pth/temp/vgg16_cifar10_act/vgg16_act.pth.tar'