Comments (8)
说实话我不知道你这样做的意义哦,为什么不直接用mobilenet呢?
- hyper-params不好说,肯定要调的
- 你给的信息比较有限, loss震荡可能原因很多,可能是lr大了。
from mobilenet.
@Zehaos 因为我这个网络不是用于分类,用作人体关键点检测,这个网络其实是一个级联的结构,分为多个stage,每个stage中包含7个convlayer,stage其实是一样的,只不过下一个stage会用到本次stage的结果,类似一个残差学习的过程。我这样做的目的是想看看mobileconv这种方式能否很好的逼近原始conv的结果,而且我觉得这样从后向前改动,比直接把所有的conv全部改成mobileconv的形式容易训练。这个网络的底座是vgg,之后级联3个stage。您觉得我是可以直接把vgg改成mobilenet,然后把所有stage都改成mobileconv直接scratch训练吗?
from mobilenet.
@sunguanxiong 噢噢,mobilenet网络设计上跟vgg是一个套路,应该可以直接把vgg换成mobilenet。
另外,关键点检测需要比较fine的特征,你可以加一些skip-connection或者deconv(或者直接跑多几个stage)。
from mobilenet.
@sunguanxiong 有tf版本的realtime pose estimation?
from mobilenet.
@Zehaos 我用的caffe版本的,mobilenet也是caffe版本的
from mobilenet.
刚好在 YOLO 上做了几个实验:
1.将 tiny-yolo-voc 前10层保持不变,后面的卷积用 depthwise conv 替换,训练 6W 轮,batchsize 64;
Recall 下降 6%,卷积层速度提升:CPU 8x 左右,GPU 感觉不到提升。模型9M,原64M。
2.将tiny-yolo-voc 卷积层全部换成DWConv,没pre-trained model,直接训练,LOSS一直震荡,
看到生成的model 才 800K,遂弃。
from mobilenet.
@Sunmyfong detection 这些应用,还是要用pretrained model才行
from mobilenet.
@Sunmyfong 可以先小lr warm up 一两个epoch,再train,不会震荡就是好事情。
from mobilenet.
Related Issues (20)
- Test single image HOT 3
- How to run in terminal?
- Mobile Net Detection Reshape Problem
- Poor performance on large objects?
- Depthwise convolution
- mobilenet_v1_eval.py has a big bug? HOT 3
- how to set the train and val datasets folder name.
- Do we need to preprocess, i.e. resize images to a fixed predefined image input resolution (for example 256*256) on our own, in order to train properly?
- Using eval_image_classifier.py HOT 1
- during classification ,I am doing performance testing on AWS with inception model flask api with gunicorn (creating multiple process) Error: OOM when allocating tensor with shape[800,1280,3] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc [[Node: Cast = CastDstT=DT_FLOAT, SrcT=DT_UINT8, _device="/job:localhost/replica:0/task:0/device:GPU:0"]] Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.
- has anyone tried the mobileNet on KITTI dataset HOT 10
- Loss does not converge HOT 4
- how to train my own classifier with mobilenet
- mobilenet not work HOT 1
- 超参数设置问题
- consul upgrade election error
- What's the meaning of the parameter: group ?
- how to usa VGG to train 448*448 size picture
- 这个官方提供的slim练习代码吗?
- 文献公式解释
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 mobilenet.