search

layumi / Person_reID_baseline_pytorch

:bouncing_ball_person: Pytorch ReID: A tiny, friendly, strong pytorch implement of person re-id / vehicle re-id baseline. Tutorial 👉https://github.com/layumi/Person_reID_baseline_pytorch/tree/master/tutorial
https://www.zdzheng.xyz
MIT License
4.12k stars 1k forks source link

Training on MSMT #277

Open Zarmeena1468 opened 3 years ago

Zarmeena1468 commented 3 years ago

Hi! I want to train the resnet50 on MSMT17 data set but i am unable to understand the preparation process. In your read me file, it says "for MSMT17, there are more than 10 cameras. You need to modify the prepare.py and test.py to read the double-digit camera ID." The problem is MSMT17 and Market have different format of data. i.e. Market has four folders of data while MSMT17 Contains only training and testing data. please guide me regarding this! Thank you

layumi commented 3 years ago

Thanks @Zarmeena1468 I just upload the https://github.com/layumi/Person_reID_baseline_pytorch/blob/master/prepare_MSMT.py You may take a look at this.

Zarmeena1468 commented 3 years ago

Thank you so much. I tried to run setup,py but it returned the error. ImportError: /home/.conda/envs/tr/lib/python3.6/site-packages/torch/lib/libmkldnn.so.0: undefined symbol: cblas_sgemm_alloc I've installed pytorch version 1.0.0 and torchvision version 0.2.1

layumi commented 3 years ago

mkl version may be wrong. You may re-install your mkl.