Comments (12)
you need to increase learning rate when you increase number of gpus
from py-r-fcn-multigpu.
thx. More concrete situation, if i use 8gpus, lr should be 8x compared with 1gpu (same iter_size) ?
from py-r-fcn-multigpu.
that worked for me, but may not always be true
from py-r-fcn-multigpu.
got it. In your coco branch, it seems that lr is still set to 1e-3 for training, while the stepsize have been set to 90000. I mean the settings in models/coco/ResNet-101/rfcn_end2end/solver_ohem.prototxt.
from py-r-fcn-multigpu.
I just created this repo for multi-gpu training and it was meant for 2 GPUs with 1 iter_size on PASCAL. But I suppose, step down would be too early for coco for that. Probably I did not optimize parameters for coco when I created this repo.
The soft-nms repo contains the training schedule for ms-coco which gets 35.1 mAP, where lr is set to 0.008. But again, its dataset specific and specific to 8 GPUs.
from py-r-fcn-multigpu.
I'll update this repo also in a month or so, so that master has all the features.
from py-r-fcn-multigpu.
awesome soft-nms repo. R-FCN in this repo got 30.8%, while soft-nms repo got 33.9%. i see that one difference between them is test set. COCO 2014 vs 2015 minival (but i think 2015 minival is same as 2014minival). and another difference is psroipooling. soft-nms use align psroipooling(proposed in mask-rcnn). does align pspooling improve 3.1%? i would like to reproduce the results given in soft-nms.
from py-r-fcn-multigpu.
It is not completely due to mask-rcnn's roi align. I implemented what I could understand from the paper and I was seeing around 1% improvement by fixing the alignment issue. I also reduced the RPN min size from 32 to 16. Training was done till 160k iterations. Probably training longer would help more. In my experience, test-dev gives 0.2% more for R-FCN, so you should get 35.3 on test-dev.
from py-r-fcn-multigpu.
thanks a lot.
from py-r-fcn-multigpu.
I will try to reproduce soft-nms experiments.
from py-r-fcn-multigpu.
I also reduced the RPN min size from 32 to 16,
does this refer to the parameters __C.TRAIN.RPN_MIN_SIZE
and __C.TEST.RPN_MIN_SIZE
?
It looks like it went from 16 to 8, not from 32 to 16.
Am i right?
from py-r-fcn-multigpu.
yes
from py-r-fcn-multigpu.
Related Issues (20)
- Can't parse message of type "caffe.NetParameter" error HOT 1
- The error when using coco dataset
- Problem of multi-GPU "'NCCL' has no attribute 'new_uid'" HOT 5
- How to train on multiple gpus?
- when use the ZF model, not the VGG16. The error happens
- GPU memory not release after interrupted the training script HOT 3
- Provide Resnet-152 training network scripts
- what is the difference between master branch and the coco branch?
- Cannot allocate memory
- How to download imagenet pretrained model HOT 1
- Can't parse message of type "caffe.NetParameter"
- Problem of multi-GPU "'NCCL' has no attribute 'new_uid'" HOT 1
- Is that possible to totally control the randomness? HOT 1
- how to solve the overflow encountered runtimewanring ?
- How to use this OHEM for Faster R-CNN?
- about coco branch, about Results on MS-COCO
- caffe version
- caffe version
- Train on coco got empty gt_box and crash
- VOC resnet50 error prototxt
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 py-r-fcn-multigpu.