Comments (3)
It is expected that the accuracy drops with ViT on STL10-40 due to the checkpoint used is from MAE. Our log also shows the performance degradation during training of STL10.
But if you are using WRN in classic setting, I think the performance should increase during training (or at least the first half of training.)
from semi-supervised-learning.
How did you train the pretrained weight? what is MAE
?
from semi-supervised-learning.
Fixed in 69a2afc
from semi-supervised-learning.
Related Issues (20)
- New feature request: Hope can compatible with Apple M1 / M2 chip. HOT 2
- Question: How to feed the model with tabular data? HOT 2
- Possibility to eval on one single data HOT 3
- Aim repository path is hard-coded HOT 1
- Regarding the bug that model_best.pth failed to save HOT 4
- About warmup does not seem to take effect HOT 3
- Checkpointing based on other evaluation metrics HOT 4
- Reduce the complexity of dependencies HOT 2
- Tensorboard log is only updated when the model is evaluated
- [Question] Regarding the Unused m Parameter in the RandAugment Class. HOT 2
- The problem with train_sampler HOT 3
- Resume Aim tracking when resuming training
- Consult about the custom network usage in the USB HOT 2
- Differences between the original paper and "results\classic_cv.csv". HOT 3
- Question about the weights in Softmatch
- "Does your algorithm support semi-supervised segmentation tasks? HOT 4
- Evaluation of STL10 (USB_CV) HOT 1
- Hard coding the dataset path HOT 5
- Unable to run eval.py on USB NLP Datasets HOT 3
- The audio dataset urbansound8k test is abnormal HOT 4
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 semi-supervised-learning.