请教加载Resnet50-ir报错“invalid load key, '\xff'.”问题

[GitHubLuGeng]

使用3.1 Experiments of SOTA backbones下的模型时，发现有些模型可以加载成功有些会报错
我的pytorch为1.1.0 GPU版本

按照如下方式加载MobileFaceNet模型成功
model = MobileFaceNet(512, 7, 7)
model = nn.DataParallel(model)
model.load_state_dict(torch.load(‘model.pt', False)

但是按照同样方式加载Resnet50-ir模型报错“invalid load key, '\xff'.”
model = Resnet(152, 0.4)
model = nn.DataParallel(model)
model.load_state_dict(torch.load(‘model.pt’, False)

请问一下要怎么样才能正确加载Resnet50-ir模型？

[wang21jun]

Please refer to backbone/backbone_def.py
model = Resnet(50, 0.4, 'ir', 512, 7, 7)

[GitHubLuGeng]

Please refer to backbone/backbone_def.py
model = Resnet(50, 0.4, 'ir', 512, 7, 7)

1.已经修改成model = Resnet(50, 0.4, 'ir', 512, 7, 7),还是报同样的错误
2.我又尝试了一下下面你们提供的加载方式，依然报错“invalid load key, '\xff'.”
backbone_conf_path = "/home/lu/PycharmProjects/FaceX-Zoo/test_protocol/backbone_conf.yaml"
backbone_factory = BackboneFactory("ResNet", backbone_conf_path)
model_loader = ModelLoader(backbone_factory)
model = model_loader.load_model(self.cfg['model_file_path'])
详细报错信息如下
Start to load the face recognition model...
backbone param:
{'depth': 50, 'drop_ratio': 0.4, 'net_mode': 'ir', 'feat_dim': 512, 'out_h': 7, 'out_w': 7}
Failed to load face recognition model.
invalid load key, '\xff'.

Process finished with exit code 255
3.我的环境是ubuntu16+python3.6+pytorch1.1.0

[GitHubLuGeng]

@wang21jun
好像是Resnet50-ir模型文件有问题，我又下载了一个Resnet152-irse的模型文件，使用同样的方式能加载成功
详细日志如下
Start to load the face recognition model...
backbone param:
{'depth': 152, 'drop_ratio': 0.4, 'net_mode': 'ir_se', 'feat_dim': 512, 'out_h': 7, 'out_w': 7}
Success!
face model loaded