Comments (3)
稀疏训练中只是对bn层的gamma进行L1正则化,达到压缩效果,但是beta没有进行压缩,它经过激活函数后会输入到下一层卷积,并且对下一层卷积的输出有影响,如果剪掉了某个通道,就会导致由beta引起的激活偏移值,并且参与剪枝的层和通道越多,产生的偏差越大,直接影响了模型的输出,所以需要将偏移值传递到下一层的bn running mean或者conv bias中。这个问题在shortcut中更加复杂,需要进行连续的传递。
from yolov3-channel-and-layer-pruning.
看稀疏情况的,有时0.01比0.1剪枝后的精度更高,它可能对应了更合理的结构,去掉了冗余的甚至是不利于精度的权重,可以下载我分享的cfg和权重剪剪试下。你也可以观察下剪枝后的cfg,哪些位置被剪的多,哪些剪得少。
from yolov3-channel-and-layer-pruning.
在SlimYOLOv3中提到“在本项目中还对激活偏移值添加了处理,降低剪枝时的精度损失”,这个怎么理解
from yolov3-channel-and-layer-pruning.
Related Issues (20)
- Can I use this to prune model MobileNetv2-Yolov4
- ValueError: not enough values to unpack (expected 2, got 1)
- 关于稀疏训练和剪枝的阈值
- 进行yolov4稀疏化训练 HOT 3
- 关于其他数据集标注格式要进行的代码修改
- 为什么稀疏训练和剪枝都要将yolov3赋给darknet?
- What are the flags -sr ,--s 0.001 ,--prune 1 --global_percent flag while doing pruning
- 刚开始推理时,GPU内存占用持续增加
- 训练时候出现一个问题 soft rratio是0测试时候 P R mAP F1也是nan这种情况怎么办呢
- The P R F1-score and [email protected] are around 90% after sparse training (--s 0.001 --prune 0). Still the sparse model is not performing well on the test dataset.
- 通道剪枝的时候遇到问题~ HOT 1
- 请问支持分组卷积的枝剪吗 HOT 1
- data文件夹下的.shapes文件需要自己准备吗?
- ValueError: Number of rows must be a positive integer, not 4.0 HOT 3
- 剪枝改进后的模型
- RuntimeError: shape '[128, 64, 3, 3]' is invalid for input of size 7067 训练v3报错
- 可以正常剪枝,但是微调并没有提升map,且微调后的bn_weight的大部分值仍接近0?
- How to transfor the pt model to ONNX format
- 在windows下没有生成val.shapes
- 没有models.yolo
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 yolov3-channel-and-layer-pruning.