Comments (5)
After some research, the first problem has been resolved. The default behaviour of ViT.forward() changed in different version of timm. When global_pool=''
, the backbone returns x[:, 0]
in timm=v0.5.x, while it returns x
in timm=v0.6.7.
from sdat.
Hi @swift-n-brutal are you now able to get the correct accuracy for VisDA dataset?
from sdat.
I can get a close result (89.8%) of CDAN+MCC+SDAT on VisDA, but met a strange behaviour. As shown in the image below, the validation accuracy (not the mAP) keeps going down as the training proceeds, and the best result (mAP 89.9%) is achieved only at the first epoch. Then I wondered whether the pretrained model was problematic. I tested two models: vit_g ('https://storage.googleapis.com/vit_models/augreg/B_16-i21k-300ep-lr_0.001-aug_medium1-wd_0.1-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.01-res_224.npz') from timm=0.5.x and vit_jx ('https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-vitjx/jx_vit_base_p16_224-80ecf9dd.pth') from timm=0.4.9. For vit_g the acc goes down, while for vit_jx the acc increases but the final mAP is (88.6%) much lower than the former one.
from sdat.
I can get a close result (89.8%) of CDAN+MCC+SDAT on VisDA, but met a strange behaviour. As shown in the image below, the validation accuracy (not the mAP) keeps going down as the training proceeds, and the best result (mAP 89.9%) is achieved only at the first epoch. Then I wondered whether the pretrained model was problematic. I tested two models: vit_g ('https://storage.googleapis.com/vit_models/augreg/B_16-i21k-300ep-lr_0.001-aug_medium1-wd_0.1-do_0.0-sd_0.0--imagenet2012-steps_20k-lr_0.01-res_224.npz') from timm=0.5.x and vit_jx ('https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-vitjx/jx_vit_base_p16_224-80ecf9dd.pth') from timm=0.4.9. For vit_g the acc goes down, while for vit_jx the acc increases but the final mAP is (88.6%) much lower than the former one.
Have you found the reason? Is it a flaw in the model or a problem with our operation?
from sdat.
@Wangzs0228 It is almost sure that the smoothness regularization is beneficial to transferability, robustness, generalization ability, etc. For a specific task, the results may vary. I am not working on this task recently. You can try the experiments to see if the results match your expectations.
from sdat.
Related Issues (12)
- How to get the accuracy reported in Paper HOT 5
- Newcomers ask for help
- Question about correctness of Domain Accuracy HOT 2
- about domain acc HOT 1
- Unavailable dataset HOT 2
- How to get correct accuracy? HOT 1
- How to get result of ResNet 101 on Visda-2017 dataset? HOT 2
- How to get result of ResNet 50 on office-home dataset? HOT 2
- Learning rate HOT 2
- Validation data is same as test data? HOT 7
- How to add SDAT on Object detection task? HOT 1
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 sdat.