PyPi release

README.md

@@ -38,6 +38,12 @@ pipeline so users can see the effect immediately.
 pip install Leakly
 ```
 
+For notebook environments that need the optional notebook dependencies:
+
+```bash
+pip install "Leakly[notebook]"
+```
+
 For the current GitHub checkout:
 
 ```bash
@@ -88,9 +94,7 @@ After label permutation, there should be no real biological, clinical, or statis
 
 A leaky pipeline may still score well if information from the full dataset enters the analysis before the train/test split or outside the cross-validation loop. Common sources include feature selection, scaling, imputation, covariate adjustment, dimensionality reduction, or hyperparameter tuning performed on all samples.
 
-This is especially problematic in high-dimensional data, such as neuroimaging, omics, or biomarker studies, where random label-specific patterns can appear meaningful by chance. If the test set influences preprocessing or feature selection, the model may preserve these random patterns and show inflated performance.
-
-Leakly tests this directly: does the pipeline still perform above chance when labels are meaningless? If yes, it does not pinpoint the exact leakage source, but it strongly indicates that the pipeline needs inspection.
+This is especially problematic in high-dimensional data, such as neuroimaging, omics, or biomarker studies, where random label-specific patterns can appear meaningful by chance. If the test set influences preprocessing or feature selection, the model may "remember" these random patterns and show inflated performance.
 
 **How many permutations should I run?**
 

example.ipynb

@@ -30,20 +30,22 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Install Leakly locally, or from GitHub when running in Colab.\n",
+    "# Install Leakly from this checkout locally, or from PyPI elsewhere.\n",
     "import importlib.metadata as metadata\n",
+    "from pathlib import Path\n",
     "import subprocess\n",
     "import sys\n",
     "\n",
-    "install_args = [sys.executable, \"-m\", \"pip\", \"install\", \"-q\", \"--upgrade\"]\n",
-    "try:\n",
-    "    import google.colab  # type: ignore  # noqa: F401\n",
-    "    install_target = \"git+https://github.com/DeMONLab-BioFINDER/Leakly.git@main\"\n",
-    "    install_args.append(\"--force-reinstall\")\n",
-    "except ImportError:\n",
-    "    install_target = \".[notebook]\"\n",
+    "repo_root = Path.cwd()\n",
+    "local_checkout = (repo_root / \"pyproject.toml\").exists() and (repo_root / \"leakly\").is_dir()\n",
     "\n",
-    "subprocess.check_call([*install_args, install_target])\n",
+    "install_args = [sys.executable, \"-m\", \"pip\", \"install\", \"-q\"]\n",
+    "if local_checkout:\n",
+    "    install_args.extend([\"-e\", \".[notebook]\"])\n",
+    "else:\n",
+    "    install_args.extend([\"--upgrade\", \"Leakly\"])\n",
+    "\n",
+    "subprocess.check_call(install_args)\n",
     "\n",
     "for module_name in list(sys.modules):\n",
     "    if module_name == \"leakly\" or module_name.startswith(\"leakly.\"):\n",
@@ -169,10 +171,10 @@
     "from leakly import permute_label\n",
     "from tqdm.auto import tqdm\n",
     "\n",
-    "# Use 2000 permutations for now to replicate `assets/AUC.png`.\n",
-    "# Use 100 permutations for a quick test run.\n",
+    "# Use 100 permutations for a quick example run.\n",
+    "# Increase to 2000 if you want to replicate `assets/AUC.png` more closely.\n",
     "# Adjust based on your computational resources and needs.\n",
-    "N_PERMUTATIONS = 2000\n",
+    "N_PERMUTATIONS = 100\n",
     "\n",
     "leakage_permuted_scores = []\n",
     "nonleakage_permuted_scores = []\n",

pyproject.toml

@@ -0,0 +1,74 @@
+[build-system]
+requires = ["setuptools>=77", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "Leakly"
+dynamic = ["version"]
+description = "Leakage checks for machine-learning pipelines using permutation tests."
+readme = "README.md"
+requires-python = ">=3.10"
+license = "MIT"
+license-files = ["LICENSE"]
+authors = [
+    { name = "DeMONLab-BioFINDER" },
+]
+keywords = [
+    "data-leakage",
+    "machine-learning",
+    "permutation-test",
+    "scikit-learn",
+]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Science/Research",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3 :: Only",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Topic :: Scientific/Engineering",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+]
+dependencies = [
+    "joblib>=1.3,<2.0",
+    "matplotlib>=3.7,<4.0",
+    "numpy>=1.23,<2.4",
+    "pandas>=1.5,<3.0",
+    "PyYAML>=6.0,<7.0",
+    "scikit-learn>=1.6,<1.7",
+    "scipy>=1.9,<1.16",
+    "tqdm>=4.64,<5.0",
+]
+
+[project.optional-dependencies]
+dev = [
+    "build>=1.0",
+    "pytest>=8.0",
+    "pytest-cov>=5.0",
+    "twine>=5.0",
+    "wheel>=0.41",
+]
+notebook = [
+    "ipykernel>=6.0",
+    "ipywidgets>=8.0",
+    "jupyter>=1.0",
+]
+
+[project.urls]
+Homepage = "https://github.com/DeMONLab-BioFINDER/Leakly"
+Source = "https://github.com/DeMONLab-BioFINDER/Leakly"
+Issues = "https://github.com/DeMONLab-BioFINDER/Leakly/issues"
+
+[tool.setuptools.dynamic]
+version = { attr = "leakly.__version__" }
+
+[tool.setuptools.packages.find]
+include = ["leakly", "leakly.*"]
+
+[tool.setuptools.package-data]
+leakly = ["*.yaml"]
+
+[tool.setuptools]
+include-package-data = true