feat: update libraries and python version; bump package version to 1.1.0

Author: diejdablju

requirements.txt

@@ -1,4 +1,5 @@
-pandas~=1.2.1
-numpy~=1.20.3
-matplotlib~=3.4.2
-rulekit~=1.6.0
+pandas >= 1.5.0, < 2.3.0
+numpy ~= 1.26.4
+matplotlib ~= 3.8.3
+rulekit ~= 1.7.6
+lifelines ~= 0.28.0

rulexai/__init__.py

@@ -1 +1 @@
-__version__ = '1.0.0'
+__version__ = '1.1.0'

rulexai/explainer.py

@@ -507,7 +507,7 @@ def __init__(
         if type == "regression":
             model = RegressionModel(model = model, feature_names= X.columns, label_name=label_name)
         elif type == "survival":
-            model = SurvivalModel(model=model, fetures_names = X.columns, survival_status_name=label_name)
+            model = SurvivalModel(model=model, feature_names = X.columns, survival_status_name=label_name)
         else:
 
             model = ClassificationModel(model = model, feature_names = X.columns, class_names = np.unique(y), label_name = label_name)

rulexai/models.py

@@ -1,4 +1,4 @@
-from rulekit.operator import BaseOperator
+from rulekit._operator import BaseOperator
 import numpy as np
 from .rule import Rule, CompoundCondition, ElementaryCondition
 from rulekit import RuleKit
@@ -484,9 +484,9 @@ def _map_rule_from_RuleKit(self, rule):
 
             compoundCondition.add_subcondition(elementaryCondition)
 
-        consequence_att, consequence_value = consequence.split(" = ")
-        consequence_val = consequence_value[1:-1]
-        consequence = ElementaryCondition(consequence_att, consequence_val, column_index=self.column_indexes[attribute])
+        consequence_att = "survival_curve"
+        consequence_val = ""
+        consequence = ElementaryCondition(consequence_att, consequence_val, column_index=0)
 
         return Rule(compoundCondition, consequence)
 

rulexai/rule.py

@@ -1,7 +1,8 @@
 import enum
 import numpy as np
 from typing import List
-np.warnings.filterwarnings('ignore')
+import warnings
+warnings.filterwarnings('ignore')
 
 # creating enumerations using class
 class LogicalOperator(enum.Enum):

setup.py

@@ -10,7 +10,7 @@
 
 setuptools.setup(
     name="rulexai",
-    version="1.0.0",
+    version="1.1.0",
     author="Dawid Macha",
     author_email="[email protected]",
     description="RuleXAI is a rule-based aproach to explain the output of any machine learning model. It is suitable for classification, regression and survival tasks.",
@@ -30,12 +30,13 @@
         "Intended Audience :: Science/Research",
     ],
     include_package_data=True,
-    python_requires=">=3.6",
+    python_requires=">=3.9",
     install_requires=[
-        "pandas ~= 1.2.1",
-        "numpy ~= 1.20.3",
-        "matplotlib ~= 3.4.2",
-        "rulekit ~= 1.6.0",
+        "pandas >= 1.5.0, < 2.3.0",
+        "numpy ~= 1.26.4",
+        "matplotlib ~= 3.8.3",
+        "rulekit ~= 1.7.6",
+        "lifelines ~= 0.28.0"
     ],
     test_suite="tests",
 )

tests/test_conditions_importances.py

@@ -14,7 +14,6 @@
 from pathlib import Path
 from rulexai.explainer import RuleExplainer
 
-
 class TestConditionsImportances(unittest.TestCase):
 
     @classmethod
@@ -103,6 +102,7 @@ def test_regression(self):
             induction_measure=Measures.C2,
             pruning_measure=Measures.C2,
             voting_measure=Measures.C2,
+            mean_based_regression = False
         )
         reg.fit(x, y)