ibcnservices / genesim Goto Github PK

• Do an evaluation with oracle method (LIME?) where predictions of the black box are used as labels to train a new decision tree. Compare GENESIM with this oracle method for Neural Nets, XGB and RF.

• Re-run all evaluations and put the accuracy of the initial ensemble (that gets merged by GENESIM) in the table as well (how much of the accuracy remains after GENESIM?)

Bro, I am a computer graduate student and I am studying your paper and would like to try to reproduce it (date:10 Dec 2023). However, unfortunately I'm currently having problems configuring my environment, and I was wondering if I could ask you to help me see if I can fix it😂🤞.
My OS is Windows 10 and I can't run install.sh directly, so I copied the commands and executed them one by one, but there is an error when installing the "rpy2" package, I tried to skip this step but I can't execute the "example.py", I would like to ask if it's possible to remove this package or to find an alternative, here is the error message.

pip list info

(venv) E:\study\Python_Workspace\GENESIM-master>pip list
Package             Version
------------------- ------------
contourpy           1.1.1       
cycler              0.12.1      
fonttools           4.46.0      
graphviz            0.20.1      
imbalanced-learn    0.11.0      
importlib-resources 6.1.1       
joblib              1.3.2       
scikit-learn        1.3.2
scipy               1.10.1
setuptools          69.0.2
six                 1.16.0
threadpoolctl       3.2.0
tzdata              2023.3
wheel               0.41.2
xgboost             2.0.2
zipp                3.17.0

Minimize dependency??It is not so easy to run install.sh successfully!

In the ISM, inTrees and GENESIM class, TreeConstructors are given along as a parameter to the merging functions. Then in these functions, decision trees are constructed using these TreeConstructors. The code would look nicer if instead a list of Decision Trees was already given along as a parameter.

Remove as many for-loops as possible and vectorize as much as possible

https://github.com/DEAP/deap

Experiment with different genetic flows (e.g. ant colony, ... / currently the most basic & generic one)

After following the instructions to setup the GENESIM, I was trying to run it for just one dataset ( wine dataset). But I was getting exceptions when the CART algorithm tries to build the trees.
Following is the snippet that i get:

CART
Traceback (most recent call last):
  File "example.py", line 65, in <module>
    clf = algorithms[algorithm].construct_classifier(train, feature_cols, label_col)
  File "D:\GENESIM\GENESIM\constructors\treeconstructor.py", line 237, in construct_classifier
    shuffle=True, random_state=None))
  File "D:\GENESIM\GENESIM\constructors\treeconstructor.py", line 336, in get_best_cart_classifier
    tree = cart.construct_classifier(train_tune, X_train_tune.columns, label_col, param_opt=False)
  File "D:\GENESIM\GENESIM\constructors\treeconstructor.py", line 249, in construct_classifier
    self.dt.fit(self.X, self.y)
  File "C:\Users\Naveen Kaushik\Anaconda2\lib\site-packages\sklearn\tree\tree.py", line 739, in fit
    X_idx_sorted=X_idx_sorted)
  File "C:\Users\Naveen Kaushik\Anaconda2\lib\site-packages\sklearn\tree\tree.py", line 199, in fit
    % self.min_samples_split)
ValueError: min_samples_split must be at least 2 or in (0, 1], got 1

Before CART it built trees with xgboost (Although with exception related to Bayesian Optimization).

What could be the possible reason for this?