Comments (4)
Hello @tanulsingh,
Thanks for opening this issue.
At the moment the only LRschedulers that you can use are ones that don't depend on the current epoch results (so ReduceOnPlateau won't work). You can use any LR that decays in a pre-defined manner.
I know this is something that we should improve, this should come with the callbacks method I think #123
So for now I suggest you to switch to torch.optim.lr_scheduler.StepLR
or torch.optim.lr_scheduler.CosineAnnealingLR
or any scheduler that does not need information about current metrics.
from tabnet.
Hey @Optimox I believe I can make those lr schedulers work by inheriting from the tabnet model base class and working on that . BTW I was able to successfully use custom loss with Tabnet . Now I plan on using lr schedulers and TPU's with tabnet . If I am successful I will update the results here . Also I was able to achieve 0.1620 on public lb in trends competition which is State of the art without any fine tuning , I will be releasing a public kernel today and I plan to finetune and use every ounce of juiceof tabnet on trends as I believe it has got a lot Of potential . I also wanted to nominate myself for abhishek's book I don;t know how to do that though
from tabnet.
@tanulsingh there is no fundamental reason not to be able to use custom loss or lr schedulers, it's just that the implementation does not give easy access to this yet.
If you spend some time changing the code to enable some functionnalities don't hesitate to create a PR so that I can review and potentially merge it.
Very happy to see that you are able to achieve good results with tabnet.
In order to be eligible to Abhishek's ebook the only things to do is to open an issue and put Abhishek's eBook
label on it, I'll add the label to this issue so you don't have anything to do.
from tabnet.
Hey @Optimox I believe I can make those lr schedulers work by inheriting from the tabnet model base class and working on that . BTW I was able to successfully use custom loss with Tabnet . Now I plan on using lr schedulers and TPU's with tabnet . If I am successful I will update the results here . Also I was able to achieve 0.1620 on public lb in trends competition which is State of the art without any fine tuning , I will be releasing a public kernel today and I plan to finetune and use every ounce of juiceof tabnet on trends as I believe it has got a lot Of potential . I also wanted to nominate myself for abhishek's book I don;t know how to do that though
@tanulsingh I don't think you can use tabnet with TPUs yet, unless you have found a way to do it already see my discussion #129 ;)
from tabnet.
Related Issues (20)
- Loss goes to -inf HOT 1
- The mask tensor M in script tab_network.py needs to be transformed to realize the objective stated in the paper: "γ is a relaxation parameter – when γ = 1, a feature is enforced to be used only at one decision step".
- Current version on conda-forge is 4.0 while 4.1 is already released HOT 8
- Minimal working example for TabNetRegressor/Classifier HOT 4
- Transfer learning, capability to change structure of model HOT 1
- Generate Embeddings for Tabular Data HOT 1
- TabNet overfits (help wanted, not a bug) HOT 9
- TabNetRegressor vs other networks HOT 1
- spike in memory when training ends HOT 8
- Severe overfitting HOT 18
- OOM problem when I search hyperparameters with Tabnet HOT 3
- Support for complex-valued datasets HOT 4
- Different classification variables in the test set and train set HOT 1
- Struggling to get model to fit - Help Wanted HOT 7
- Optimizing TabNet for Disease Classification with Continuous Audio Features HOT 1
- Interpreting Sparsity on Global Importance HOT 5
- ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all() HOT 1
- Validation loss HOT 1
- Lightweight Fine-tunning or few-shot learning for limited labeled data HOT 1
- Maybe `drop_last` should be set as False in default? 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 tabnet.