Remove Python 3.7 and 3.8

.github/workflows/push-pull.yml

@@ -7,7 +7,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ["3.7", "3.8", "3.9", "3.10", "3.11"]
+        python-version: ["3.9", "3.10", "3.11"]
     steps:
       - uses: actions/checkout@v3
       - name: Set up Python ${{ matrix.python-version }}

pyproject.toml

@@ -13,44 +13,21 @@ repository = "https://github.com/DataDome/sliceline"
 documentation = "https://sliceline.readthedocs.io/en/stable/"
 
 [tool.poetry.dependencies]
-python = ">=3.7, <3.12"
-numpy = [
-    { version = "^1.21", python = "<3.9" },
-    { version = "^1.25", python = ">=3.9" },
-]
-scikit-learn = [
-    { version = "^1", python = "<3.8" },
-    { version = "^1.3", python = ">=3.8,<3.9" },
-    { version = "^1.4", python = ">=3.9" },
-]
-scipy = [
-    { version = "^1", python = "<3.9" },
-    { version = "^1.12", python = ">=3.9" },
-]
+python = ">=3.9, <3.12"
+numpy = "^1.25"
+scikit-learn = "^1.5.0"
+scipy = "^1.12"
 
 [tool.poetry.dev-dependencies]
-black = [
-    { version = "^23", python = "<3.8" },
-    { version = "^24", python = ">=3.8" },
-]
-flake8 = [
-    { version = "^5", python = "<3.8.1" },
-    { version = "^6", python = ">=3.8.1" },
-]
+black = "^24"
+flake8 = "^6"
 jupyter = "^1.0.0"
-matplotlib = [
-    { version = "^3", python = "<3.8" },
-    { version = "^3.7", python = ">=3.8,<3.9" },
-    { version = "^3.8", python = ">=3.9" },
-]
+matplotlib = "^3.8"
 nbconvert = "^6.5.0"
 optbinning = "^0.15.0"
-pandas = [
-    { version = "^1", python = "<3.9" },
-    { version = "^1.5", python = ">=3.9" },
-]
-pytest = "^6.2.5"
-pytest-benchmark = "^3.4.1"
+pandas = "^1.5"
+pytest = "^7.2.0"
+pytest-benchmark = "^4.0.0"
 pytest-cov = "^3.0.0"
 Sphinx = "^4.0.0"
 sphinx-rtd-theme = "^1.0.0"

sliceline/slicefinder.py

@@ -1,6 +1,7 @@
 """
 The slicefinder module implements the Slicefinder class.
 """
+
 import logging
 from typing import Tuple, Union
 
@@ -426,7 +427,8 @@ def _get_pruned_s_r(
         self, slices: sp.csr_matrix, statistics: np.ndarray
     ) -> Tuple[sp.csr_matrix, np.ndarray]:
         """Prune invalid slices.
-        Do not affect overall pruning effectiveness due to handling of missing parents."""
+        Do not affect overall pruning effectiveness due to handling of missing parents.
+        """
         valid_slices_mask = (statistics[:, 3] >= self.min_sup) & (
             statistics[:, 1] > 0
         )

sliceline/validation.py

@@ -330,7 +330,6 @@ def check_array(
     estimator=None,
     input_name="",
 ):
-
     """Input validation on an array, list, sparse matrix or similar.
 
     By default, the input is checked to be a non-empty 2D array containing

tests/conftest.py

@@ -3,6 +3,7 @@
 - basic_test_data
 - experiments
 """
+
 import numpy as np
 import pytest
 from scipy import sparse as sp
@@ -35,7 +36,7 @@ def basic_test_data():
     ).T
 
     # error or label of elements
-    # By default, the algorithm identifies slices targeting 0 of E.
+    # By default, the algorithm identifies slices targeting 1 of errors.
     errors = np.array([1, 1, 1, 1, 0, 0, 0, 0])
 
     slicefinder_model = slicefinder.Slicefinder(

tests/experiment.py

@@ -1,6 +1,7 @@
 """
 The experiment module implements the Experiment dataclass.
 """
+
 from dataclasses import dataclass
 from typing import Dict, List, Union
 

tests/test_slicefinder.py

@@ -2,6 +2,7 @@
 The test_slicefinder module tests slicefinder class and methods.
 Tests are run on different Experiments.
 """
+
 import numpy as np
 import pytest
 from scipy import sparse as sp