wujiyang

Yes，margin is the last fc layer corresponding to the number of classes and it's not needed in evaluation.

@vanlong96tg Sorry，currently I can't share my pretrained model. You can use the code to train a well-performanced one.

train_list 就是一个分类数据的list，每个类别后面给一个类别编号即可，你只要训练过分类模型应该就知道

@lymanblue All the training and validation/test data have been aligned by MTCNN to the size of 112*112

@lymanblue The cleaned-MS1M I used is provided by DeepGlint, it only has 3.9M images, while InsightFace has a 5.8M cleaned version. The dataloader of MS1M is as same as the...

You can use your own data (MS1M-V2) to train the models directly.

No, train.rec and train.idx are the mxnet's data format, the dataloader I provided is suitable for images,just like this: 00000 --- 00000-00001.jpg --- 00000-00002.jpg --- 00000-00003.jpg 00001 --- 00001-00001.jpg ---00001-00002.jpg...

Well, you can use it to parse the train.rec file to get original images.

sorry， I did this mini project during my campus time, and now I don't have the train/val/test data.

The training epoch is not fixed, it depends on your total size of training images and your batch size. You can choose an appropriate one for yourself. As for others...