Shayan Aqabarary

> Hi there, I want to ask how to train in multi-datasets? For example, I want to train in market1501 and dukemtmc, then test in cuhk03. Would I just set...

> Hi，your work is really work for me, but In model zoo, some pretrained models aren't provided, like osnet_ibn_0_5, how can i train it by myself? thank you for you...

> Hi, After a few search, I find out triplet loss more effective than softmax loss. I read this two articles: 1) https://www.franciscosalg.com/triplet-loss/ 2) https://neptune.ai/blog/how-to-choose-loss-function-for-face-recognition I test the original code...

> hello, I also can't get the paper experience result, do you have to get it? I really want to know the parameter settings that can get the same results,...

> loss: name: 'triplet' softmax: label_smooth: True > > train: optim: 'amsgrad' lr: 0.0015 max_epoch: 250 batch_size: 64 fixbase_epoch: 0 open_layers: ['classifier'] lr_scheduler: 'cosine' > > ``` > cfg.loss.triplet.weight_t =...

> Thanks for your great job! Would you please show me the performance of OSNet with reranking ? As you konw , in many cases , the rank1 and mAP...

> Hi, You should change sources and targets (changing targets is optional) in this type: sources=['market1501', 'cuhk03', 'dukemtmc-reid','msmt17','viper','grid','cuhk01','sensereid','ilids','prid2011','cuhk02','cuhksysu'], targets=['market1501', 'cuhk03', 'dukemtmc-reid','msmt17','viper','grid','cuhk01','sensereid','ilids','prid2011','cuhk02','cuhksysu'] You can read this link for more information: https://kaiyangzhou.github.io/deep-person-reid/user_guide.html#combine-multiple-datasets

> Thanks for the reply. I have followed the instructions with no success... Going CUHK03 class, dataset_url is None. during runtime it fails in line 36 `self.download_dataset(self.dataset_dir, self.dataset_url)` but since...