zhaixingzi
zhaixingzi
There is a useless parameter beta_ ? why?
Do i still need to set kv_store in using mxnet? why?
Hi, there is a similar idea in "Correlation Congruence for Knowledge Distillation arXiv:1904.01802v1". Do you think which is more efficent ?
您好，您用的pickle模块的版本号是多少？ 我在执行这一句时 `self.data_dict = pickle.load(file)` 遇到错误 ·UnicodeDecodeError: 'gbk' codec can't decode byte 0x80 in position 0: illegal multibyte sequence·
训练出的模型的效果如何？
作者您好，请问训练出的模型的效果如何？和mxnet的代码相比，模型精度有损失吗？
hello, i complie your code on ubuntu14.04 to finetune the wu's model with scaled_cosine_softmax. your caffe could be compiled successfully.But ,when i go to finetune the model with four nvidia...
大神您好。aligned_lfw.zip 中矫正好的图片您是用什么方法得到的呢？ 我在重现ydwen的实验时，我用自己矫正的lfw数据集做测试，accuracy只有0.981,而用您矫正的LFW做测试时accuracy有0.988。 我使用的是mtcnn做的人脸检测和关键点检测。
which version of pytorch is used in this project?
字体大小能设置吗？ 登录时能设置代理吗？