How to load load_state_dict for ResNet model

[Ramchandra_Cheke]

# create new OrderedDict that does not contain `module.`
from collections import OrderedDict

# original saved file with DataParallel
ckpt = torch.load("pretraied_model.pth")
state_dict = ckpt['model']    # incase there are extra parameters in the model

new_state_dict = OrderedDict()

for k, v in state_dict.items():
    name =  k.replace("module.encoder.","encoder.")
    new_state_dict[name] = v

# load params
model.encoder.load_state_dict(new_state_dict,strict=False)

Thanks to post

[solved] KeyError: 'unexpected key "module.encoder.embedding.weight" in state_dict'

I was thinking about something like the following: # original saved file with DataParallel state_dict = torch.load('myfile.pth.tar') # create new OrderedDict that does not contain `module.` from collections import OrderedDict new_state_dict = OrderedDict() for k, v in state_dict.items(): name = k[7:] # remove `module.` new_state_dict[name] = v # load params model.load_state_dict(new_state_dict)