mobilefacenet-V2

mobilefacenet-V2 copied to clipboard

Can you share parameter file (.params)? I cant load pretrained model (-symbol.json) without it.

2

quocnhat

how to find the peretrain models

4

how to find the peretrain models,i cannot find in this: https://github.com/aidlearning/AidLearning-FrameWork/tree/master/examples/facencnn cannot find weight files?

walkingmanroad

Why is acc always zero

2

I used train_softmax,py in https://github.com/deepinsight/insightface/blob/master/src/train_softmax.py The parameters are set like this： CUDA_VISIBLE_DEVICES='0' python -u train_softmax.py --network y1 --ckpt 2 --loss-type 4 --lr 0.00001 --emb-size 512 --per-batch-size 80 --wd 0.0000001 --data-dir...

2151330

Thank you very much for your work. Could you give me the link to the test dataset,such as lfw,agedb_30......

2151330

The reason why the accuracy is so high

1

Hello, you can train such a good model, what parameters do you mainly adjust, can you share your experience?

zhangxiaopang88

why the pretrained model with low accuracy

6

Why I use your training model with low initial accuracy? Am I wrong? cmd:CUDA_VISIBLE_DEVICES='0' python -u train_softmax.py --network y1 --ckpt 2 --loss-type 4 --lr 0.01 --lr-steps 40000,60000,70000 --wd 0.00004 --fc7-wd-mult...

lly8752

emb-size

2

why your emb size is 512? the result is unfair compared to the orginal paper.and the model size is too large

cavalleria

As the title , can you tell me the difference between yours and mobilefacenet? Some details，such as LeakyReLU and wd extremely grateful |

AlphaQi

Problem with the module symbol_utils

1

I suggest you add symbol_utils.py to your code. Thank you so much.

tauruswcc