-
Notifications
You must be signed in to change notification settings - Fork 19
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Is it suitable for regression prediction? #7
Comments
Hi @775269512, I think it is intuitive to use SDT on regression tasks, simply change the training criterion in |
hi, I did a simple experiment. Although the overall loss is decreasing, the output of each sample is the same and cannot be regressed (here out_dim = 1), x = tensor([[ 1., 1., 1., 1., 1.], i got this result I don't know how to change it. It seems that when outdim = 1, the value of each leaf node is the same. QAQ |
Could you show me the code snippet on training and evaluating? |
yep, it's here. In addition, I found that a paper is based on SDT, and I will study it: "SDTR: Soft Decision Tree Regressor for Tabular Data". It is difficult to understand the differentiable decision tree. This is a good interpretable model and I want to use it to do something. btw, I will also study postgraduate in nju next semester. I find you are my senior~ '''
''' |
It looks like you are using the full batch training process (i.e., without using a dataloader that samples batches), maybe you should |
Hello, I'd like to ask if I want to make regression prediction and output_ Dim = = 1, is SDT applicable (it seems to be only used for classification model?)
Thanks!
The text was updated successfully, but these errors were encountered: