Comments (8)
你好 经过我们很努力的尝试 可以稍微提高 但离 30 还是差得很远,因为我们是二值网络 注重于降低计算量和显存,因为性能方面会有一定受损
您或许可以关注一下最近的工作 比如这篇,根据他们的 Table 3, 在 COCO 上达到了 30 左右的性能
from bidet.
好的。非常感谢博士您的回复。
我现在就是想找一个二值网络,我的目标也是想找一个降低计算量和显存的网络,看在 COCO 上能否达到30左右(或较近)的Map。您说的这篇论文,我会仔细去阅读,遗憾的是他未公开源码,而我的复现能力菜的抠脚,也不知道能否复现。
祝好
from bidet.
博士,您好
我还有一个疑问, 您设计的这个全部都是二值计算,还是除了第一层和最后一层是全精度的,其他都是二值计算
祝好
from bidet.
你好 依照二值网络的惯例 第一和最后一层是全精度 只有中间是二值
很好理解 因为输出层要预测 bbox coordinate, 这是一个连续的值,因此如果仍然用二值输出层 只能得到整数预测,误差必然很大
from bidet.
是的,我之前尝试过将其他网络改成这样(第一和最后一层是全精度,中间层为二值),但与全精度计算的结果相差就比较大,然后搜索资料时发现您的论文与代码,感觉还比较OK,然后就开始训练测试。可惜的是我还没复现到您论文的效果,您当时训练花了多久,然后用的啥平台呀。
from bidet.
年代过于久远... 应该就是正常 2张 (?) GTX1080Ti 训要训几天 具体不记得了 大概 3 天?
from bidet.
请问是8G显存吗?
from bidet.
不是,查了一下,是12G显存。是说我的8G小卡训练起来batchsize设置为8,num_worker=1时,训练起来都为何这么吃力呢
祝好
from bidet.
Related Issues (20)
- bidet测试效果不好 HOT 1
- loss为inf HOT 1
- test error HOT 1
- params HOT 3
- Vgg arch in SSD implementation is different from original vgg HOT 2
- 数据集划分问题 HOT 10
- 关于计算量和参数量 HOT 2
- 关于训练 HOT 2
- 关于二值化带来的参数缩减 HOT 4
- 关于检测头的二值化 HOT 6
- 关于检测头二值化问题的请教 HOT 7
- 关于faster rcnn的FPN HOT 1
- Resnet18的layer中存在未二值化的卷积 HOT 2
- 在其他数据集上训练 HOT 1
- faster rcnn训练路径
- VOC数据集结果复现
- IB准则
- None of the weights are binarized HOT 1
- 预训练 HOT 8
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 bidet.