Comments (11)
博主,我重新训练了,换的resnet,现在的准确率挺高的,模型也收敛到3左右
from build-your-own-face-model.
正常吧应该【笑哭】
你的数据集有多大?如果数据集很大,考虑是欠拟合问题,用 resnet 版的,不要用 mobilenet 版的。你给的信息太少不知道怎么回你【笑哭】
from build-your-own-face-model.
数据集确实挺大的,用的是celeba数据集,然后我应该是没有改动你上面说的这两个版本的。
from build-your-own-face-model.
你试过 resnet 版的,也是很低吗?可以训练久一点试试。
from build-your-own-face-model.
from build-your-own-face-model.
你试过 resnet 版的,也是很低吗?可以训练久一点试试。
我训练的比较久,但是效果一样,我接下来实施你说的这个版本
from build-your-own-face-model.
我现在改了,明天应该会有结果,我训练好了给你反馈吧,感谢博主
from build-your-own-face-model.
博主,我重新训练了,换的resnet,现在的准确率挺高的,模型也收敛到3左右
你好,我最近也在复现centerface,请问这个resnet版本的应该如何修改?
from build-your-own-face-model.
博主,我重新训练了,换的resnet,现在的准确率挺高的,模型也收敛到3左右
您好,请问如何获得resnet版本需要的资料(网络模型等)
from build-your-own-face-model.
我的也是不收敛,用的是CeleBa数据集,1万个分类,请教m和s超参多少啊?我看了您的代码,用的还是默认的m=0.5和s=30,不过我看了原始论文,他推荐是m=1.35,大概是77度。另外,论文里没有推荐半径s的值。这2个超参您在训练过程中,如何把握的呢?请教 :https://github.com/piginzoo/arcface-pytorch.git
from build-your-own-face-model.
mobilenet用了博主指定的数据集:CASIA-WebFace-Clean训练,lfw-align-128为测试,,,训练50轮,在5-6轮后, loss后面都在10-5-10之间跳,但是ACC稳定在0.96*上下,,,用resnet训练了几轮了,没一点收敛的迹象。这样正常吗?哪里设置有问题呢?
from build-your-own-face-model.
Related Issues (20)
- LFW RGB Image vs Gray Scale Image HOT 2
- IDE HOT 1
- Training Issue HOT 1
- Face detector bounding box issue HOT 2
- 关于arcface损失函数的问题 HOT 4
- 人脸检测 特征点heatmap是否有错? HOT 1
- 在自己的数据集上训练 HOT 1
- 请问您测试过检测模型在widerface al 上面的map吗? HOT 1
- 请问人脸识别模型的label是one-hot形式的吗? HOT 1
- fmobile训练一个epoch要多久 HOT 3
- How to run the demo HOT 2
- 图像分类top1
- Google drive link
- 关于导入权重的问题 HOT 1
- 精度达不到24.pth HOT 1
- arcface的实现问题 HOT 1
- train问题 HOT 2
- 如何用于人脸分类任务的训练和测试?
- 单人图像身份识别
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from build-your-own-face-model.