Comments (3)
@bruAristimunha Thanks for reporting the issue. This sounds serious. We are always careful not to break backwards compatibility, and if we do are very explicit about it and provide a guide for upgrading. So this sounds like something slipped through.
I think the best way forward would be to identify the exact commit that broke your tests. That's probably best achieved by using git bisect
. Let me know if you want to give it a try, I'll support as best as I can. Alternatively, if you could give me a small script (small in the sense that it runs fast) that fails with the current skorch version but worked previously, I can bisect on my own.
from skorch.
Hi @BenjaminBossan!
Thank you very much for opening this issue. I followed his recommendation, and it was unclear why we now need the classes
parameter when we tested in Python version 3.8 or higher. I tried to follow the instructions and create a mini-code to deliver to you; however, as we are going to make a code sprint next week, I preferred to incorporate this new parameter into the examples in the braindecode rather than finding out exactly what generated this requirement.
From what I studied, it could even have been sklearn. The lesson has been learned, and we will improve the control of these dependecies. For me, we can close this issue, and I thank you for your immense work on skorch :)
from skorch.
From what I studied, it could even have been sklearn
This is very much possible. Actually, some skorch code also started failing because somewhere inside of sklearn, there was a change that suddenly required classes_
when previously it was not required. See this comment:
It seems that the classes_ attribute is now required on classifiers for some stuff related to sklearn scoring, where previously it wasn't
I also tried to figure out the exact change but then gave up and just accepted it.
I thank you for your immense work on skorch :)
Thanks a lot, always happy to hear it.
from skorch.
Related Issues (20)
- IndexError: invalid index of a 0-dim tensor. Use `tensor.item()` in Python or `tensor.item()` in C++ to convert a 0-dim tensor to a number HOT 8
- Enable using a generator as data loader
- Question: weird valid loss when re-scaling y
- ReadTheDocs: Wrong theme of docs
- How to tune number of epochs?
- Issues with deployment script
- dill load and sklearn clone result in error HOT 7
- how to integrate pytorch-tabnet into skorch framework? HOT 2
- ImportError: cannot import name 'uses_placeholder_y' from 'skorch.dataset' HOT 1
- Saving and Loading not working HOT 2
- Add slides for Pydata Amsterdam HOT 3
- model.history not save output of all epochs HOT 2
- I can't use gpu cuda tensor with NeuralNet HOT 1
- Permit to pass '**predict_params' to 'predict' method as for 'fit' method HOT 2
- Skorch forwarding data columns as kwargs when using gridsearchcv HOT 4
- Skorch weird handling of input data HOT 3
- LLM caching breaks with shared-prefix labels under certain conditions HOT 3
- Activating (deactivating) callbacks at specific epochs or milestones and SequentialLR HOT 1
- Dictionary Input and Custom Collate Function 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 skorch.