model's dimension doesn't match checkpoint's dimension?

[chithize]

HI,参照myTest目录下的video测试demo脚本,
下载页面提供的模型链接文件
将 154行 ：File = os.path.join(exp_args.model_root, 'model_best.pth.tar')
改成
bestModelFile = os.path.join(exp_args.model_root, 'mobilenetv2_total_with_prior_channel.tar')
运行后产生如下错误

videotest.py:101: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
cf = load(cont)
finish load config file ...
Traceback (most recent call last):
File "videotest.py", line 158, in
netmodel_video.load_state_dict(checkpoint_video['state_dict'])
File "/home/hcruise/.local/lib/python2.7/site-packages/torch/nn/modules/module.py", line 487, in load_state_dict
.format(name, own_state[name].size(), param.size()))
RuntimeError: While copying the parameter named stage0.0.weight, whose dimensions in the model are torch.Size([32, 4, 3, 3]) and whose dimensions in the checkpoint are torch.Size([32, 3, 3, 3]).

这是什么情况，我是安装主页上说的安装的pytorch 0.3.0post4, 及torchvision 0.2.0
为什么load模型后又说自己的模型dimension不对呢？