Facial-Expression-Recognition.Pytorch
Facial-Expression-Recognition.Pytorch copied to clipboard
A CNN based pytorch implementation on facial expression recognition (FER2013 and CK+), achieving 73.112% (state-of-the-art) in FER2013 and 94.64% in CK+ dataset
Thank you for sharing your nice work! I really enjoyed the results. (it tends to output `sad` face 🤔) I run this script on MacBook, so I Update the demo...
What is the use of AvgPool2D at the end in vgg.py ?
Pleasee help me, i can use this code for my rasberry pi 3
萌新提问
请问 mainpro_FER.py 执行完成之后,可以这样保存模型到文件吗? torch.save(net, '80.pth') 我训练了80代,另外再次加载模型时,怎么把一张正常的人脸图片传入模型呢?目前我把一个人脸图片裁剪为48*48,但是推理结果的正确度不好
有问题的小伙伴,可以私聊我,我已经完成基本项目的复现,邮箱[email protected]
I see you build the resnet18 model, why not use the resnet18 model torchversion.model provided and just modify the linear layer. Thank you so much!
运行preprocess_fer2013后,得到的输出结果为(0,),请问这个问题该如何解决?是什么原因导致的呢? 
_, term_width = os.popen('stty size', 'r').read().split() ValueError: not enough values to unpack (expected 2, got 0) Can you tell me how to fix the problem?
混淆矩阵的运行
