ankonzoid / learningx Goto Github PK

Hi,

Is it possible to use a subset of the features for the tree splits and the rest for the leaf models?

Thanks,
C

Hello @ankonzoid
Your implemetation is really good. I was wondering if we could add some more visualization part, so that we can see the agent moving in the grid as it learns.
Please suggest how can I do that or some references?
Thanks

Thank you for your wonderful repo! I am learning how to use your Logistic Model Trees, but I find an error in the impelement of Logistic Regression.

In dvanced_ML/model_tree/models/logistic_regr.py, the functionlogistic_regr.prediction() is trying to call np.ones(), but package numpy didn't be imported inlogistic_regr.py. It seems we should add import numpy as np in this script.

Hello I'm pretty new in data science and machine learning and i came a across your model tree which I find really cool. I just want to ask will it work with N dimensional train data? Since the example is with 1D train data and I am not sure if i can answer this by myself with looking at the code with my current knowledge.

I am not sure how difficult it would be to implement this. Is it possible to add a feature that would allow one to regress in the leaves only on a subset of features? E.g. if we have 3 features f1, f2, f3, the xgb trees are constructed by using features f1, f2 (or all of them) but the linear regression is done on features f2, f3 (notice the overlap).

Thanks.

This is a great package. Nice work! Could we please make this pip installable? It massively simplifies dependency management when working in production settings. Thank you!

I was using your ModelTree(great work by the way!!), and I would like to know if there is a way to visualize the model tree as curves on a graph, similar to the images in your readme?