search

dreamquark-ai / tabnet

PyTorch implementation of TabNet paper : https://arxiv.org/pdf/1908.07442.pdf
https://dreamquark-ai.github.io/tabnet/
MIT License
2.61k stars 485 forks source link

BUG: Pretraining example does not work #425

Closed canbooo closed 2 years ago

canbooo commented 2 years ago

Describe the bug

pretraining example does not work due to possibly outdated keywords n_shared_decoder and n_indep_decoder

What is the current behavior? The notebook does not run

If the current behavior is a bug, please provide the steps to reproduce.

  1. Run the pretraining notebook
  2. See the TypeError: __init__() got an unexpected keyword argument 'n_shared_decoder' at cell 10.

Expected behavior

Example notebooks should run without error and should not require modification

Other relevant information: poetry version: 1.1.13 python version: 3.9.10 Operating System: Linux Additional tools:

Optimox commented 2 years ago

The code works, that's why we have CI.

The problem is probably that you are running the develop branch notebook with the master branch code.

Either look at the master branch notebook or use the code from development branch.

By default the github page is in the develop branch, which can be quite confusing.

canbooo commented 2 years ago

You are right, my bad. In my defence, it is weird confusing to have develop as the default branch.