search

BookerDeWitt / MSBDN-DFF

The source code of CVPR 2020 paper "Multi-Scale Boosted Dehazing Network with Dense Feature Fusion"
334 stars 69 forks source link

model.pkl unpacking problem #6

Open zsjhhh opened 4 years ago

zsjhhh commented 4 years ago

Hi, I want to take a look at what's in model.pkl. However, when I open it using the following code, the output is a series of number.

import pickle info = pickle.load(open('./model/MSBDNet/checkpoint/msbdn_model.pkl', 'rb'), encoding='utf-8') print(info)

Output: 119547037146038801333356

Is that the decoding problem? Or do I need some specific package to unpack it, if so，could you please tell which one should I use? Thank you very much.

adityac8 commented 4 years ago

Hi @zsjhhh If you have a look at the README, it states that model.pkl is the checkpoints file. This means that it contains the weights of the model architecture. You can load it similar to the line given in test.py.

model = torch.load(test_pkl, map_location=lambda storage, loc: storage)

However, looking at the contents of it won't give you any information. Thanks

hello-trouble commented 3 years ago

hello, sorry to bother you. when I load the pretrained model in another project like this: model = torch.load(test_pkl, map_location=lambda storage, loc: storage), the error infomation is "ModuleNotFoundError: No module named 'networks'" , I am puzzled about why the structure should be the same as the original project, with MSBDN-DFF-v1-1.py in the networks directory. Can you help me about this problem? Thank you for your help in advance.