-
you continue train for 5000 iterations, the training set is LFW dataset or the original training phase dataset?
i am confused about the training dataset. can you explain to me?
-
This result mostly come from the modification of center loss. Instead of updating centers manually according the paper, I made centers as learnable variable and applied a little bit weight decay to it…
-
From the form of the loss function, I think adding the lambda can be seen as using smaller m. I have plotted the curve of `lambda=5, m=4` and find that it is approximately m=1.5.
 whi…
-
Why we don't use whole training dataset in each epoch? There is example for casia to use 1000 iterations (epoch size) with batch size 90 and it is 1/5 of whole casia set per epoch.
I would like to …
-
When I try Classifier training of inception resnet v1 as described on the Wiki my validation rate is always very low (max of about 0.5). I have tried several things, I tried upping the epochs, upping …
-
您在这里的结果对照您的论文来看应该是最新版的在微软数据集训练的结果，从这里可以看出在LFW上pytorch的精度略低于caffe的精度。
我只使用casia webface作为训练集 使用您的pytorch训练代码（lightcnn-9） 最终在LFW 6000对的评测上仅能达到100% - EER 97.2 TPR@FAR=1% 94.333 BLUFR 结果为82.67 …
-
Hi @AlfredXiangWu ,
my training datafile like this:
/home/jonanza/datasets/casia/CASIA__144/3599667/037.png 0
/home/jonanza/datasets/casia/CASIA__144/1466221/082.png 1
/home/jonanza/datasets/cas…
-
many thx for the previous reply！
two more issues：
1. have you tested the model on the CASIA_B dataset？ can you achieve the 90% around accuracy？
2. have you observed the final performance dif…