Comments (6)
That such accuracy drop is kind of weird. I did not meet such things before.
- Have you checked the training loss and accuracy? if things go wrong, training loss would become very large.
- It may relate to bn issue in pytorch. Have you followed readme to "Disable cudnn for batchnorm layer to solve bug in pytorch0.4.0".
from fast_human_pose_estimation_pytorch.
Thanks for your reply!
- And actually, for s4b1, the validation loss would be very large but at that time training loss is normally I think, the validation loss will be larger after several epochs, and the validation accuracy has been dropped to almost zero already. See below:
Epoch LR Train Loss Val Loss Train Acc Val Acc
1.000000 0.000250 0.002626 0.009073 0.161942 0.291490
2.000000 0.000250 0.002116 0.009197 0.300657 0.397661
3.000000 0.000250 0.001956 0.004193 0.399708 0.420880
4.000000 0.000250 0.001839 0.067800 0.472296 0.016799
5.000000 0.000250 0.001757 9.869279 0.520801 0.000000
6.000000 0.000250 0.001726 678.551853 0.542042 0.007716
7.000000 0.000250 0.001753 3865.202707 0.544255 0.000042
8.000000 0.000250 0.001758 37764.363059 0.538761 0.000000
9.000000 0.000250 0.001917 933.443237 0.410885 0.000068
- I forgot to do that step you mentioned, I will try to disable cudnn for the bn layer, thanks!
from fast_human_pose_estimation_pytorch.
And one more question, what do num_feature and inplane really influence? looks like they are the channel variables of the first several layers, but do they affect the accuracy or just affect the size of models? Thanks!
from fast_human_pose_estimation_pytorch.
I don't think the network structure has such big impact. The training acc looks good, while validation acc goes to zero. It looks like related to bn layer who behaves differently in train and eval mode. I suspect it has something to do with bn issue I mentioned.
from fast_human_pose_estimation_pytorch.
Thanks for your proposals! it is actually the BN layer eval bug on pytorch0.4.0, I disable the cudnn for batchnorm layer, the val loss and val acc seems normal then.
By the way, do you think if I use s8b1 teacher model train s4b2 student, and use s4b2 as the teacher train s1b2, will get better accuracy for s1b2 than directly use s8b1 teacher model train s1b2 student? That means steps KD. Thanks!
from fast_human_pose_estimation_pytorch.
Your idea looks interesting, but not sure if this performs better or not. KD is a kind of regulariation.
Ensemble(s8b1, s4b2 ..) to teach s1b2
Iterative KD, s8b1->s4b2->s1b2 .
Good luck!
from fast_human_pose_estimation_pytorch.
Related Issues (20)
- hi, does it support multi-person pose estimation? HOT 1
- **Loss: inf | Acc: 0.0000**while evaluation from mpii.py HOT 6
- A simple question about the paper HOT 1
- requeset.txt need pip install torch, but i use conda ,use pytorch canot run?who use conda to run this/ HOT 1
- Leeds jason file?
- Student model overfits really early training HOT 4
- train at mpii,acc so small HOT 8
- where is the code of teacher to student?? HOT 5
- Why the accuracy differs a lot in the experiment and paper? HOT 1
- License HOT 2
- check the result of demo with the image sample.jpg
- python example/mpii_kd.py -a hg --stacks 2 --blocks 1 --checkpoint checkpoint/hg_s2_b1_mobile/ mobile=True --teacher_stack 8 --teacher_checkpoint HOT 1
- train my dataset HOT 1
- unlabel dataset HOT 3
- loss function HOT 3
- 有个问题,试了下其他类型图片,效果并不好,是否只适用于mpii数据集上图片? HOT 2
- RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same HOT 1
- mobile=false HOT 3
- Unable to export to onnx HOT 3
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 fast_human_pose_estimation_pytorch.