resnet50 pretrain model: size mismatch

[Kewenjing1020]

HI, i'm trying to use your code to reproduce the experiments in your paper. I use the given command line and pretrained float model, while it reports the error of size mismatching:

=> loading checkpoint '/home/wenjing/model-best.pth.tar'
Traceback (most recent call last):
File "main.py", line 685, in
main()
File "main.py", line 201, in main
load_model(model, checkpoint)
File "main.py", line 121, in load_model
model.load_state_dict(new_state_dict, strict=False) #strict false in case the loaded doesn't have alll variables like running mean
File "/home/kewenjing/anaconda3/envs/pytorch-0.4.1-py3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 719, in load_state_dict
self.class.name, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for ResNet_imagenet:
size mismatch for layer1.0.conv2.weight: copying a param of torch.Size([64, 2, 3, 3]) from checkpoint, where the shape is torch.Size([64, 64, 3, 3]) in current model.
size mismatch for layer1.1.conv2.weight: copying a param of torch.Size([64, 2, 3, 3]) from checkpoint, where the shape is torch.Size([64, 64, 3, 3]) in current model.
size mismatch for layer1.2.conv2.weight: copying a param of torch.Size([64, 2, 3, 3]) from checkpoint, where the shape is torch.Size([64, 64, 3, 3]) in current model.
size mismatch for layer2.0.conv2.weight: copying a param of torch.Size([128, 4, 3, 3]) from checkpoint, where the shape is torch.Size([128, 128, 3, 3]) in current model.
size mismatch for layer2.1.conv2.weight: copying a param of torch.Size([128, 4, 3, 3]) from checkpoint, where the shape is torch.Size([128, 128, 3, 3]) in current model.
size mismatch for layer2.2.conv2.weight: copying a param of torch.Size([128, 4, 3, 3]) from checkpoint, where the shape is torch.Size([128, 128, 3, 3]) in current model.
size mismatch for layer2.3.conv2.weight: copying a param of torch.Size([128, 4, 3, 3]) from checkpoint, where the shape is torch.Size([128, 128, 3, 3]) in current model.
size mismatch for layer3.0.conv2.weight: copying a param of torch.Size([256, 8, 3, 3]) from checkpoint, where the shape is torch.Size([256, 256, 3, 3]) in current model.
size mismatch for layer3.1.conv2.weight: copying a param of torch.Size([256, 8, 3, 3]) from checkpoint, where the shape is torch.Size([256, 256, 3, 3]) in current model.
size mismatch for layer3.2.conv2.weight: copying a param of torch.Size([256, 8, 3, 3]) from checkpoint, where the shape is torch.Size([256, 256, 3, 3]) in current model.
size mismatch for layer3.3.conv2.weight: copying a param of torch.Size([256, 8, 3, 3]) from checkpoint, where the shape is torch.Size([256, 256, 3, 3]) in current model.
size mismatch for layer3.4.conv2.weight: copying a param of torch.Size([256, 8, 3, 3]) from checkpoint, where the shape is torch.Size([256, 256, 3, 3]) in current model.
size mismatch for layer3.5.conv2.weight: copying a param of torch.Size([256, 8, 3, 3]) from checkpoint, where the shape is torch.Size([256, 256, 3, 3]) in current model.
size mismatch for layer4.0.conv2.weight: copying a param of torch.Size([512, 16, 3, 3]) from checkpoint, where the shape is torch.Size([512, 512, 3, 3]) in current model.
size mismatch for layer4.1.conv2.weight: copying a param of torch.Size([512, 16, 3, 3]) from checkpoint, where the shape is torch.Size([512, 512, 3, 3]) in current model.
size mismatch for layer4.2.conv2.weight: copying a param of torch.Size([512, 16, 3, 3]) from checkpoint, where the shape is torch.Size([512, 512, 3, 3]) in current model.