Problem with pre-trained models

[anilsh]

I am not able to open the downloaded tar files. Anyone else faced similar issue?

[lsongx]

@anilsh open it with torch.load(), no need to use tar xf. sorry for the confusing suffix

[mittlin]

Use torch.load() BUT got UnicodeDecodeError: 'ascii' codec can't decode byte 0x91 in position 0: ordinal not in range(128). any idea??

[lsongx]

@mittlin Which file are you dealing with?

[YMing-Li]

Use torch.load() BUT got UnicodeDecodeError: 'utf-8' codec can't decode byte in position 0. any idea??

[lsongx]

@YMing-Li Could you try loading with PyTorch 0.3 / Python 2?

[YMing-Li]

@YMing-Li Could you try loading with PyTorch 0.3 / Python 2?

Thanks a lot. I am now encountering a new problem. After I run source_train.py directly, the mAP is only 53%, which is far worse than the trained network given. The small bugs in source_train.py have been fixed. Do some parameters need to be changed? My version : Python3.7.4 and Pytorch1.3.1

[lsongx]

@YMing-Li Maybe some params need to be changed. Im not sure. As an alternative, you could try https://github.com/open-mmlab/OpenUnReID for your environment.