diff --git a/.github/actions/setup-python-env/action.yml b/.github/actions/setup-python-env/action.yml new file mode 100644 index 0000000..1c95877 --- /dev/null +++ b/.github/actions/setup-python-env/action.yml @@ -0,0 +1,27 @@ +name: "Setup Python Env" +description: "Install project dependencies" +runs: + using: "composite" + steps: + - uses: actions/setup-python@v5 + with: + python-version: "3.12" + + - name: Show where we are + shell: bash + run: | + echo "action path: $GITHUB_ACTION_PATH" + echo "workspace: $GITHUB_WORKSPACE" + pwd + ls -la + + - name: Install dependencies + shell: bash + working-directory: ${{ github.workspace }} # <- repo root for pip + run: | + echo "CWD:" && pwd + test -f pyproject.toml && echo "Found pyproject.toml" || (echo "pyproject.toml MISSING" && exit 1) + python -m pip install --upgrade pip + pip install -r requirements.txt + pip install -r requirements_dev.txt + pip install -e . \ No newline at end of file diff --git a/.github/workflows/mypy.yml b/.github/workflows/mypy.yml new file mode 100644 index 0000000..02e0442 --- /dev/null +++ b/.github/workflows/mypy.yml @@ -0,0 +1,13 @@ +name: mypy + +on: + - push + - pull_request + +jobs: + typecheck: + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - uses: ./.github/actions/setup-python-env + - run: mypy . --python-executable "$(command -v python)" \ No newline at end of file diff --git a/.github/workflows/pytest.yml b/.github/workflows/pytest.yml new file mode 100644 index 0000000..4f6a709 --- /dev/null +++ b/.github/workflows/pytest.yml @@ -0,0 +1,12 @@ +name: pytest + +on: + [push, pull_request] + +jobs: + test: + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - uses: ./.github/actions/setup-python-env + - run: pytest tests/ diff --git a/.github/workflows/ruff.yml b/.github/workflows/ruff.yml new file mode 100644 index 0000000..fe62313 --- /dev/null +++ b/.github/workflows/ruff.yml @@ -0,0 +1,13 @@ +name: ruff + +on: + [push, pull_request] + +jobs: + test: + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - uses: ./.github/actions/setup-python-env + - run: ruff check src/* + - run: ruff check tests/* \ No newline at end of file diff --git a/.gitignore b/.gitignore index 5779ef8..6359176 100644 --- a/.gitignore +++ b/.gitignore @@ -93,3 +93,4 @@ dmypy.json # Editors .vscode/ .idea/ +.ruff_cache/ \ No newline at end of file diff --git a/README.md b/README.md index 1388c9d..d0e509f 100644 --- a/README.md +++ b/README.md @@ -3,6 +3,10 @@ [![PyPI version](https://badge.fury.io/py/RegularizedDiscriminantAnalysis.svg)](https://badge.fury.io/py/RegularizedDiscriminantAnalysis) [![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/downloads/) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) +[![mypy](https://github.com/F3z11/RegularizedDiscriminantAnalysis/actions/workflows/mypy.yml/badge.svg)](https://github.com/F3z11/RegularizedDiscriminantAnalysis/actions/workflows/mypy.yml) +[![ruff](https://github.com/F3z11/RegularizedDiscriminantAnalysis/actions/workflows/mypy.yml/badge.svg)](https://github.com/F3z11/RegularizedDiscriminantAnalysis/actions/workflows/ruff.yml) +[![pytest](https://github.com/F3z11/RegularizedDiscriminantAnalysis/actions/workflows/mypy.yml/badge.svg)](https://github.com/F3z11/RegularizedDiscriminantAnalysis/actions/workflows/pytest.yml) + A scikit-learn compatible implementation of Regularized Discriminant Analysis (RDA) as proposed by Friedman (1989). diff --git a/examples/rda_comparison.png b/examples/rda_comparison.png index 1df8823..bd77428 100644 Binary files a/examples/rda_comparison.png and b/examples/rda_comparison.png differ diff --git a/examples/rda_example.ipynb b/examples/rda_example.ipynb index d9638d8..4606a39 100644 --- a/examples/rda_example.ipynb +++ b/examples/rda_example.ipynb @@ -2,280 +2,372 @@ "cells": [ { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "id": "cdfc5c48", "metadata": {}, + "outputs": [], + "source": [ + "import os\n", + "import sys\n", + "\n", + "# Add the src directory to sys.path to allow importing the package\n", + "# In Jupyter notebooks, we use the current working directory\n", + "\n", + "notebook_dir = os.getcwd()\n", + "src_path = os.path.abspath(os.path.join(notebook_dir, '../src'))\n", + "if src_path not in sys.path:\n", + " sys.path.insert(0, src_path)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5880b2c7", + "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:11,913] A new study created in memory with name: no-name-bc74ae1e-417e-4142-99f6-7c4a176f508e\n" + "/Users/matteo/Desktop/RegularizedDiscriminantAnalysis/.venv/lib/python3.12/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n", + " from .autonotebook import tqdm as notebook_tqdm\n" ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Starting Optuna optimization for RDA...\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Best trial: 6. Best value: 0.964203: 12%|█▏ | 12/100 [00:00<00:01, 79.85it/s]" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[I 2025-11-30 12:16:11,932] Trial 0 finished with value: 0.9405499034490974 and parameters: {'lambda_': 0.3745401188473625, 'gamma': 0.9507143064099162}. Best is trial 0 with value: 0.9405499034490974.\n", - "[I 2025-11-30 12:16:11,946] Trial 1 finished with value: 0.9586308453199655 and parameters: {'lambda_': 0.7319939418114051, 'gamma': 0.5986584841970366}. Best is trial 1 with value: 0.9586308453199655.\n", - "[I 2025-11-30 12:16:11,958] Trial 2 finished with value: 0.9418120351588911 and parameters: {'lambda_': 0.15601864044243652, 'gamma': 0.15599452033620265}. Best is trial 1 with value: 0.9586308453199655.\n", - "[I 2025-11-30 12:16:11,969] Trial 3 finished with value: 0.9417086396316889 and parameters: {'lambda_': 0.05808361216819946, 'gamma': 0.8661761457749352}. Best is trial 1 with value: 0.9586308453199655.\n", - "[I 2025-11-30 12:16:11,980] Trial 4 finished with value: 0.9587561586031734 and parameters: {'lambda_': 0.6011150117432088, 'gamma': 0.7080725777960455}. Best is trial 4 with value: 0.9587561586031734.\n", - "[I 2025-11-30 12:16:11,992] Trial 5 finished with value: 0.9315313655650543 and parameters: {'lambda_': 0.020584494295802447, 'gamma': 0.9699098521619943}. Best is trial 4 with value: 0.9587561586031734.\n", - "[I 2025-11-30 12:16:12,003] Trial 6 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8324426408004217, 'gamma': 0.21233911067827616}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,014] Trial 7 finished with value: 0.9418120351588911 and parameters: {'lambda_': 0.18182496720710062, 'gamma': 0.18340450985343382}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,059] Trial 8 finished with value: 0.9478730574661374 and parameters: {'lambda_': 0.3042422429595377, 'gamma': 0.5247564316322378}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,069] Trial 9 finished with value: 0.949638374184725 and parameters: {'lambda_': 0.43194501864211576, 'gamma': 0.2912291401980419}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,081] Trial 10 finished with value: 0.9622325043112632 and parameters: {'lambda_': 0.9779960740812002, 'gamma': 0.005997182955817526}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,093] Trial 11 finished with value: 0.9622325043112632 and parameters: {'lambda_': 0.9695371107878658, 'gamma': 0.005297120813991947}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,105] Trial 12 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9826843743915114, 'gamma': 0.34386754411301484}. Best is trial 6 with value: 0.9642030548068284.\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Best trial: 6. Best value: 0.964203: 14%|█▍ | 14/100 [00:00<00:01, 79.85it/s]" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[I 2025-11-30 12:16:12,119] Trial 13 finished with value: 0.9624044302256257 and parameters: {'lambda_': 0.8191814679646803, 'gamma': 0.34514502423241655}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,132] Trial 14 finished with value: 0.9605705384957721 and parameters: {'lambda_': 0.8033861539264717, 'gamma': 0.3823219598797172}. Best is trial 6 with value: 0.9642030548068284.\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "Best trial: 18. Best value: 0.968038: 29%|██▉ | 29/100 [00:00<00:00, 73.70it/s]" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[I 2025-11-30 12:16:12,146] Trial 15 finished with value: 0.9585377606489264 and parameters: {'lambda_': 0.6233322887262935, 'gamma': 0.21996287655199753}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,161] Trial 16 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.8785249836669968, 'gamma': 0.43435627175641134}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,173] Trial 17 finished with value: 0.9587561586031734 and parameters: {'lambda_': 0.6474686234892436, 'gamma': 0.6418737837292194}. Best is trial 6 with value: 0.9642030548068284.\n", - "[I 2025-11-30 12:16:12,185] Trial 18 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9949795628716714, 'gamma': 0.10957088319196047}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,197] Trial 19 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8638006223663643, 'gamma': 0.10745332541720692}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,208] Trial 20 finished with value: 0.9585383716638844 and parameters: {'lambda_': 0.5222000238688389, 'gamma': 0.0699073149786461}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,220] Trial 21 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.86722053407673, 'gamma': 0.13873418445653635}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,232] Trial 22 finished with value: 0.9585377606489264 and parameters: {'lambda_': 0.7011625953937599, 'gamma': 0.24657915778684059}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,243] Trial 23 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.9155545015337317, 'gamma': 0.09595034968921504}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,255] Trial 24 finished with value: 0.9639755713758053 and parameters: {'lambda_': 0.7611286058853513, 'gamma': 0.07611210796030307}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,268] Trial 25 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9007511719320646, 'gamma': 0.2758040748165358}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,279] Trial 26 finished with value: 0.9626679167729257 and parameters: {'lambda_': 0.8262029988804105, 'gamma': 0.46483403323954436}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,291] Trial 27 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9962950533036381, 'gamma': 0.18805033768699958}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,303] Trial 28 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9938308060476384, 'gamma': 0.21113298038398604}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,314] Trial 29 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9961002695170315, 'gamma': 0.32719859065967455}. Best is trial 18 with value: 0.9680378848214337.\n" - ] - }, + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "import optuna\n", + "from optuna.samplers import TPESampler\n", + "from optuna.trial import Trial\n", + "from sklearn.datasets import load_breast_cancer\n", + "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis, QuadraticDiscriminantAnalysis\n", + "from sklearn.metrics import classification_report, f1_score\n", + "from sklearn.model_selection import cross_val_score, train_test_split\n", + "from sklearn.preprocessing import StandardScaler\n", + "\n", + "from regularizeddiscriminantanalysis import RegularizedDiscriminantAnalysis" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "2b49367e", + "metadata": {}, + "outputs": [], + "source": [ + "# Load and preprocess data\n", + "X, y = load_breast_cancer(return_X_y=True)\n", + "\n", + "# Standardize features\n", + "scaler = StandardScaler()\n", + "X = scaler.fit_transform(X)\n", + "\n", + "X_train, X_test, y_train, y_test = train_test_split(\n", + " X, y, test_size=0.3, random_state=42\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "7bb41741", + "metadata": {}, + "outputs": [], + "source": [ + "def objective(trial: Trial) -> float:\n", + " \"\"\"Objective function for Optuna optimization.\n", + " \n", + " Parameters\n", + " ----------\n", + " trial : optuna.trial.Trial\n", + " Optuna trial object for hyperparameter suggestions.\n", + " \n", + " Returns\n", + " -------\n", + " float\n", + " Mean cross-validation accuracy (to be maximized).\n", + " \"\"\"\n", + " # Suggest hyperparameters\n", + " lambda_ = trial.suggest_float('lambda_', 0.0, 1.0)\n", + " gamma = trial.suggest_float('gamma', 0.0, 1.0)\n", + " \n", + " # Create RDA model\n", + " rda = RegularizedDiscriminantAnalysis(\n", + " lambda_=lambda_,\n", + " gamma=gamma\n", + " )\n", + " \n", + " # Cross-validate on training set\n", + " scores = cross_val_score(\n", + " rda, X_train, y_train, cv=5, scoring='f1'\n", + " )\n", + " \n", + " return scores.mean()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8b409dc4", + "metadata": {}, + "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 18. Best value: 0.968038: 31%|███ | 31/100 [00:00<00:00, 73.70it/s]" + "[I 2025-12-03 15:44:34,255] A new study created in memory with name: no-name-a47ab3b2-6c90-4d31-9cc6-5cd92e130d46\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,326] Trial 30 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.9313561331022734, 'gamma': 0.5311240303514603}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,341] Trial 31 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9277530213886823, 'gamma': 0.19290690763646673}. Best is trial 18 with value: 0.9680378848214337.\n" + "Starting Optuna optimization for RDA...\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 18. Best value: 0.968038: 43%|████▎ | 43/100 [00:00<00:00, 69.86it/s]" + "Best trial: 18. Best value: 0.968038: 27%|██▋ | 27/100 [00:00<00:00, 145.64it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,353] Trial 32 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9368619896648693, 'gamma': 0.19035422620770448}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,369] Trial 33 finished with value: 0.9622770860387544 and parameters: {'lambda_': 0.718467976910326, 'gamma': 0.15497320863603303}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,382] Trial 34 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.9317974296307155, 'gamma': 0.4078226521899437}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,429] Trial 35 finished with value: 0.9548582579117297 and parameters: {'lambda_': 0.7764745448824741, 'gamma': 0.8629273218532529}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,441] Trial 36 finished with value: 0.9679408429885428 and parameters: {'lambda_': 0.9249522481741105, 'gamma': 0.02640632531968369}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,453] Trial 37 finished with value: 0.9581138207326318 and parameters: {'lambda_': 0.5499552175778062, 'gamma': 0.032237319394933364}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,465] Trial 38 finished with value: 0.9622770860387544 and parameters: {'lambda_': 0.6805714232849659, 'gamma': 0.1291859274856733}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,477] Trial 39 finished with value: 0.9489926986192438 and parameters: {'lambda_': 0.2550570692138962, 'gamma': 0.07587457593049411}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,489] Trial 40 finished with value: 0.949638374184725 and parameters: {'lambda_': 0.41951979886658475, 'gamma': 0.2725778591431674}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,501] Trial 41 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9338011080149786, 'gamma': 0.17919644347860825}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,514] Trial 42 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.928359647396847, 'gamma': 0.16315807888908807}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,525] Trial 43 finished with value: 0.9639753328432574 and parameters: {'lambda_': 0.8729560778020485, 'gamma': 0.05652471061570347}. Best is trial 18 with value: 0.9680378848214337.\n" + "[I 2025-12-03 15:44:34,269] Trial 0 finished with value: 0.9405499034490974 and parameters: {'lambda_': 0.3745401188473625, 'gamma': 0.9507143064099162}. Best is trial 0 with value: 0.9405499034490974.\n", + "[I 2025-12-03 15:44:34,277] Trial 1 finished with value: 0.9586308453199655 and parameters: {'lambda_': 0.7319939418114051, 'gamma': 0.5986584841970366}. Best is trial 1 with value: 0.9586308453199655.\n", + "[I 2025-12-03 15:44:34,284] Trial 2 finished with value: 0.9418120351588911 and parameters: {'lambda_': 0.15601864044243652, 'gamma': 0.15599452033620265}. Best is trial 1 with value: 0.9586308453199655.\n", + "[I 2025-12-03 15:44:34,291] Trial 3 finished with value: 0.9417086396316889 and parameters: {'lambda_': 0.05808361216819946, 'gamma': 0.8661761457749352}. Best is trial 1 with value: 0.9586308453199655.\n", + "[I 2025-12-03 15:44:34,297] Trial 4 finished with value: 0.9587561586031734 and parameters: {'lambda_': 0.6011150117432088, 'gamma': 0.7080725777960455}. Best is trial 4 with value: 0.9587561586031734.\n", + "[I 2025-12-03 15:44:34,303] Trial 5 finished with value: 0.9315313655650543 and parameters: {'lambda_': 0.020584494295802447, 'gamma': 0.9699098521619943}. Best is trial 4 with value: 0.9587561586031734.\n", + "[I 2025-12-03 15:44:34,309] Trial 6 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8324426408004217, 'gamma': 0.21233911067827616}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,315] Trial 7 finished with value: 0.9418120351588911 and parameters: {'lambda_': 0.18182496720710062, 'gamma': 0.18340450985343382}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,321] Trial 8 finished with value: 0.9478730574661374 and parameters: {'lambda_': 0.3042422429595377, 'gamma': 0.5247564316322378}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,327] Trial 9 finished with value: 0.949638374184725 and parameters: {'lambda_': 0.43194501864211576, 'gamma': 0.2912291401980419}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,335] Trial 10 finished with value: 0.9622325043112632 and parameters: {'lambda_': 0.9779960740812002, 'gamma': 0.005997182955817526}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,342] Trial 11 finished with value: 0.9622325043112632 and parameters: {'lambda_': 0.9695371107878658, 'gamma': 0.005297120813991947}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,349] Trial 12 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9826843743915114, 'gamma': 0.34386754411301484}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,356] Trial 13 finished with value: 0.9624044302256257 and parameters: {'lambda_': 0.8191814679646803, 'gamma': 0.34514502423241655}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,364] Trial 14 finished with value: 0.9605705384957721 and parameters: {'lambda_': 0.8033861539264717, 'gamma': 0.3823219598797172}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,370] Trial 15 finished with value: 0.9585377606489264 and parameters: {'lambda_': 0.6233322887262935, 'gamma': 0.21996287655199753}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,377] Trial 16 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.8785249836669968, 'gamma': 0.43435627175641134}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,384] Trial 17 finished with value: 0.9587561586031734 and parameters: {'lambda_': 0.6474686234892436, 'gamma': 0.6418737837292194}. Best is trial 6 with value: 0.9642030548068284.\n", + "[I 2025-12-03 15:44:34,391] Trial 18 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9949795628716714, 'gamma': 0.10957088319196047}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,398] Trial 19 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8638006223663643, 'gamma': 0.10745332541720692}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,405] Trial 20 finished with value: 0.9585383716638844 and parameters: {'lambda_': 0.5222000238688389, 'gamma': 0.0699073149786461}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,411] Trial 21 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.86722053407673, 'gamma': 0.13873418445653635}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,419] Trial 22 finished with value: 0.9585377606489264 and parameters: {'lambda_': 0.7011625953937599, 'gamma': 0.24657915778684059}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,426] Trial 23 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.9155545015337317, 'gamma': 0.09595034968921504}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,432] Trial 24 finished with value: 0.9639755713758053 and parameters: {'lambda_': 0.7611286058853513, 'gamma': 0.07611210796030307}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,439] Trial 25 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9007511719320646, 'gamma': 0.2758040748165358}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,446] Trial 26 finished with value: 0.9626679167729257 and parameters: {'lambda_': 0.8262029988804105, 'gamma': 0.46483403323954436}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,453] Trial 27 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9962950533036381, 'gamma': 0.18805033768699958}. Best is trial 18 with value: 0.9680378848214337.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 18. Best value: 0.968038: 44%|████▍ | 44/100 [00:00<00:00, 69.86it/s]" + "Best trial: 18. Best value: 0.968038: 30%|███ | 30/100 [00:00<00:00, 144.89it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,540] Trial 44 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.950077311477228, 'gamma': 0.12314126872897249}. Best is trial 18 with value: 0.9680378848214337.\n" + "[I 2025-12-03 15:44:34,460] Trial 28 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9938308060476384, 'gamma': 0.21113298038398604}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,467] Trial 29 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9961002695170315, 'gamma': 0.32719859065967455}. Best is trial 18 with value: 0.9680378848214337.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 18. Best value: 0.968038: 59%|█████▉ | 59/100 [00:00<00:00, 74.32it/s]" + "Best trial: 18. Best value: 0.968038: 56%|█████▌ | 56/100 [00:00<00:00, 142.75it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,557] Trial 45 finished with value: 0.9679832297082971 and parameters: {'lambda_': 0.794719377911135, 'gamma': 0.011072641696413843}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,572] Trial 46 finished with value: 0.9548582579117297 and parameters: {'lambda_': 0.7656622389162583, 'gamma': 0.795867761893573}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,585] Trial 47 finished with value: 0.9636007970533502 and parameters: {'lambda_': 0.8453835875986682, 'gamma': 0.000447503146362771}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,597] Trial 48 finished with value: 0.9625297435088338 and parameters: {'lambda_': 0.7931664560189527, 'gamma': 0.3003022297365558}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,611] Trial 49 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.8944096198677683, 'gamma': 0.22915558625596372}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,625] Trial 50 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8263603617860318, 'gamma': 0.1915006181584841}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,637] Trial 51 finished with value: 0.9675809164488409 and parameters: {'lambda_': 0.9552647964150446, 'gamma': 0.042298024782466546}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,650] Trial 52 finished with value: 0.9658434647113893 and parameters: {'lambda_': 0.9033569387039972, 'gamma': 0.030354712659122346}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,662] Trial 53 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9646094505623444, 'gamma': 0.11070136603774108}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,674] Trial 54 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8571898942020828, 'gamma': 0.17160742440164026}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,687] Trial 55 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9460932345838065, 'gamma': 0.24327043508793061}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,699] Trial 56 finished with value: 0.9471588068893901 and parameters: {'lambda_': 0.14726135966679627, 'gamma': 0.07995783697385485}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,710] Trial 57 finished with value: 0.9643283680900364 and parameters: {'lambda_': 0.743412514567491, 'gamma': 0.14084733985772593}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,721] Trial 58 finished with value: 0.9660041408360044 and parameters: {'lambda_': 0.888809364275138, 'gamma': 0.0018031707817141446}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,734] Trial 59 finished with value: 0.9660361774364112 and parameters: {'lambda_': 0.8132415970588371, 'gamma': 0.10312105188623803}. Best is trial 18 with value: 0.9680378848214337.\n" + "[I 2025-12-03 15:44:34,475] Trial 30 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.9313561331022734, 'gamma': 0.5311240303514603}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,482] Trial 31 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9277530213886823, 'gamma': 0.19290690763646673}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,489] Trial 32 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9368619896648693, 'gamma': 0.19035422620770448}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,496] Trial 33 finished with value: 0.9622770860387544 and parameters: {'lambda_': 0.718467976910326, 'gamma': 0.15497320863603303}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,503] Trial 34 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.9317974296307155, 'gamma': 0.4078226521899437}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,510] Trial 35 finished with value: 0.9548582579117297 and parameters: {'lambda_': 0.7764745448824741, 'gamma': 0.8629273218532529}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,517] Trial 36 finished with value: 0.9679408429885428 and parameters: {'lambda_': 0.9249522481741105, 'gamma': 0.02640632531968369}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,524] Trial 37 finished with value: 0.9581138207326318 and parameters: {'lambda_': 0.5499552175778062, 'gamma': 0.032237319394933364}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,531] Trial 38 finished with value: 0.9622770860387544 and parameters: {'lambda_': 0.6805714232849659, 'gamma': 0.1291859274856733}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,538] Trial 39 finished with value: 0.9489926986192438 and parameters: {'lambda_': 0.2550570692138962, 'gamma': 0.07587457593049411}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,545] Trial 40 finished with value: 0.949638374184725 and parameters: {'lambda_': 0.41951979886658475, 'gamma': 0.2725778591431674}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,552] Trial 41 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9338011080149786, 'gamma': 0.17919644347860825}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,560] Trial 42 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.928359647396847, 'gamma': 0.16315807888908807}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,567] Trial 43 finished with value: 0.9639753328432574 and parameters: {'lambda_': 0.8729560778020485, 'gamma': 0.05652471061570347}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,574] Trial 44 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.950077311477228, 'gamma': 0.12314126872897249}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,582] Trial 45 finished with value: 0.9679832297082971 and parameters: {'lambda_': 0.794719377911135, 'gamma': 0.011072641696413843}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,589] Trial 46 finished with value: 0.9548582579117297 and parameters: {'lambda_': 0.7656622389162583, 'gamma': 0.795867761893573}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,596] Trial 47 finished with value: 0.9636007970533502 and parameters: {'lambda_': 0.8453835875986682, 'gamma': 0.000447503146362771}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,603] Trial 48 finished with value: 0.9625297435088338 and parameters: {'lambda_': 0.7931664560189527, 'gamma': 0.3003022297365558}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,610] Trial 49 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.8944096198677683, 'gamma': 0.22915558625596372}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,617] Trial 50 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8263603617860318, 'gamma': 0.1915006181584841}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,624] Trial 51 finished with value: 0.9675809164488409 and parameters: {'lambda_': 0.9552647964150446, 'gamma': 0.042298024782466546}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,631] Trial 52 finished with value: 0.9658434647113893 and parameters: {'lambda_': 0.9033569387039972, 'gamma': 0.030354712659122346}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,638] Trial 53 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9646094505623444, 'gamma': 0.11070136603774108}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,645] Trial 54 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8571898942020828, 'gamma': 0.17160742440164026}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,652] Trial 55 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9460932345838065, 'gamma': 0.24327043508793061}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,659] Trial 56 finished with value: 0.9471588068893901 and parameters: {'lambda_': 0.14726135966679627, 'gamma': 0.07995783697385485}. Best is trial 18 with value: 0.9680378848214337.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 18. Best value: 0.968038: 60%|██████ | 60/100 [00:00<00:00, 74.32it/s]" + "Best trial: 18. Best value: 0.968038: 58%|█████▊ | 58/100 [00:00<00:00, 142.75it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,746] Trial 60 finished with value: 0.9581138207326318 and parameters: {'lambda_': 0.5839950905218905, 'gamma': 0.040677113559204406}. Best is trial 18 with value: 0.9680378848214337.\n" + "[I 2025-12-03 15:44:34,666] Trial 57 finished with value: 0.9643283680900364 and parameters: {'lambda_': 0.743412514567491, 'gamma': 0.14084733985772593}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,673] Trial 58 finished with value: 0.9660041408360044 and parameters: {'lambda_': 0.888809364275138, 'gamma': 0.0018031707817141446}. Best is trial 18 with value: 0.9680378848214337.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 62. Best value: 0.969642: 72%|███████▏ | 72/100 [00:01<00:00, 67.42it/s]" + "Best trial: 62. Best value: 0.969642: 80%|████████ | 80/100 [00:00<00:00, 142.58it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,760] Trial 61 finished with value: 0.9675809164488409 and parameters: {'lambda_': 0.9711731520804705, 'gamma': 0.04201130545919867}. Best is trial 18 with value: 0.9680378848214337.\n", - "[I 2025-11-30 12:16:12,773] Trial 62 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9508075098561362, 'gamma': 0.056244346906978646}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,828] Trial 63 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9156673942714, 'gamma': 0.18779221642120197}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,841] Trial 64 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8570285675531686, 'gamma': 0.0930356922350811}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,853] Trial 65 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9998256780269942, 'gamma': 0.1428728274756551}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,865] Trial 66 finished with value: 0.9677733906413153 and parameters: {'lambda_': 0.9431307659196638, 'gamma': 0.0650277307843502}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,877] Trial 67 finished with value: 0.9405458249423531 and parameters: {'lambda_': 0.8945919735928143, 'gamma': 0.9974698284809731}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,890] Trial 68 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.9703798049018051, 'gamma': 0.5734796019414754}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,901] Trial 69 finished with value: 0.9625297435088338 and parameters: {'lambda_': 0.7952457760714058, 'gamma': 0.21510989325553243}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,913] Trial 70 finished with value: 0.9624044302256257 and parameters: {'lambda_': 0.8391599068450151, 'gamma': 0.3127457078789469}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,925] Trial 71 finished with value: 0.9660359389038634 and parameters: {'lambda_': 0.9309919022707629, 'gamma': 0.06587394637804811}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,937] Trial 72 finished with value: 0.9679408429885428 and parameters: {'lambda_': 0.9550018731661827, 'gamma': 0.02276492597690838}. Best is trial 62 with value: 0.9696415225094471.\n" + "[I 2025-12-03 15:44:34,680] Trial 59 finished with value: 0.9660361774364112 and parameters: {'lambda_': 0.8132415970588371, 'gamma': 0.10312105188623803}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,687] Trial 60 finished with value: 0.9581138207326318 and parameters: {'lambda_': 0.5839950905218905, 'gamma': 0.040677113559204406}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,694] Trial 61 finished with value: 0.9675809164488409 and parameters: {'lambda_': 0.9711731520804705, 'gamma': 0.04201130545919867}. Best is trial 18 with value: 0.9680378848214337.\n", + "[I 2025-12-03 15:44:34,701] Trial 62 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9508075098561362, 'gamma': 0.056244346906978646}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,708] Trial 63 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9156673942714, 'gamma': 0.18779221642120197}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,715] Trial 64 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8570285675531686, 'gamma': 0.0930356922350811}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,722] Trial 65 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9998256780269942, 'gamma': 0.1428728274756551}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,729] Trial 66 finished with value: 0.9677733906413153 and parameters: {'lambda_': 0.9431307659196638, 'gamma': 0.0650277307843502}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,736] Trial 67 finished with value: 0.9405458249423531 and parameters: {'lambda_': 0.8945919735928143, 'gamma': 0.9974698284809731}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,743] Trial 68 finished with value: 0.9623079902332238 and parameters: {'lambda_': 0.9703798049018051, 'gamma': 0.5734796019414754}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,750] Trial 69 finished with value: 0.9625297435088338 and parameters: {'lambda_': 0.7952457760714058, 'gamma': 0.21510989325553243}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,757] Trial 70 finished with value: 0.9624044302256257 and parameters: {'lambda_': 0.8391599068450151, 'gamma': 0.3127457078789469}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,764] Trial 71 finished with value: 0.9660359389038634 and parameters: {'lambda_': 0.9309919022707629, 'gamma': 0.06587394637804811}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,771] Trial 72 finished with value: 0.9679408429885428 and parameters: {'lambda_': 0.9550018731661827, 'gamma': 0.02276492597690838}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,778] Trial 73 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9153839141692414, 'gamma': 0.36682754086610136}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,785] Trial 74 finished with value: 0.9679408429885428 and parameters: {'lambda_': 0.95984002189027, 'gamma': 0.021027452267466502}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,792] Trial 75 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8697118369726705, 'gamma': 0.12148338542812644}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,799] Trial 76 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9835055660816806, 'gamma': 0.0929321117384037}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,806] Trial 77 finished with value: 0.9645018085027793 and parameters: {'lambda_': 0.8821431595145821, 'gamma': 0.2687168009585018}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,813] Trial 78 finished with value: 0.9622770860387544 and parameters: {'lambda_': 0.6716352378801798, 'gamma': 0.15124109631655452}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,846] Trial 79 finished with value: 0.9557206583394695 and parameters: {'lambda_': 0.3512786202203141, 'gamma': 0.01787818127210962}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,860] Trial 80 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9998028064027679, 'gamma': 0.05997150971024942}. Best is trial 62 with value: 0.9696415225094471.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 62. Best value: 0.969642: 73%|███████▎ | 73/100 [00:01<00:00, 67.42it/s]" + "Best trial: 62. Best value: 0.969642: 82%|████████▏ | 82/100 [00:00<00:00, 142.58it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,949] Trial 73 finished with value: 0.9641418819630776 and parameters: {'lambda_': 0.9153839141692414, 'gamma': 0.36682754086610136}. Best is trial 62 with value: 0.9696415225094471.\n" + "[I 2025-12-03 15:44:34,868] Trial 81 finished with value: 0.9679040707719953 and parameters: {'lambda_': 0.9261372408718175, 'gamma': 0.055874046788384285}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,876] Trial 82 finished with value: 0.9602473841726178 and parameters: {'lambda_': 0.9792605254825869, 'gamma': 0.6795065122967671}. Best is trial 62 with value: 0.9696415225094471.\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 62. Best value: 0.969642: 88%|████████▊ | 88/100 [00:01<00:00, 73.00it/s]" + "Best trial: 62. Best value: 0.969642: 100%|██████████| 100/100 [00:00<00:00, 135.95it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:12,965] Trial 74 finished with value: 0.9679408429885428 and parameters: {'lambda_': 0.95984002189027, 'gamma': 0.021027452267466502}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,983] Trial 75 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8697118369726705, 'gamma': 0.12148338542812644}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:12,994] Trial 76 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9835055660816806, 'gamma': 0.0929321117384037}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,007] Trial 77 finished with value: 0.9645018085027793 and parameters: {'lambda_': 0.8821431595145821, 'gamma': 0.2687168009585018}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,019] Trial 78 finished with value: 0.9622770860387544 and parameters: {'lambda_': 0.6716352378801798, 'gamma': 0.15124109631655452}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,031] Trial 79 finished with value: 0.9557206583394695 and parameters: {'lambda_': 0.3512786202203141, 'gamma': 0.01787818127210962}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,043] Trial 80 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9998028064027679, 'gamma': 0.05997150971024942}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,055] Trial 81 finished with value: 0.9679040707719953 and parameters: {'lambda_': 0.9261372408718175, 'gamma': 0.055874046788384285}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,067] Trial 82 finished with value: 0.9602473841726178 and parameters: {'lambda_': 0.9792605254825869, 'gamma': 0.6795065122967671}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,079] Trial 83 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9545551175538949, 'gamma': 0.11663356357577104}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,092] Trial 84 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9938068645780518, 'gamma': 0.19352476238108154}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,104] Trial 85 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.909044947880665, 'gamma': 0.08247871692776128}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,117] Trial 86 finished with value: 0.9574284676858443 and parameters: {'lambda_': 0.4897512550751938, 'gamma': 0.02369928399171662}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,128] Trial 87 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9376702608354682, 'gamma': 0.16041658085464028}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,140] Trial 88 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9289231204022531, 'gamma': 0.1662178519485552}. Best is trial 62 with value: 0.9696415225094471.\n" + "[I 2025-12-03 15:44:34,883] Trial 83 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9545551175538949, 'gamma': 0.11663356357577104}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,890] Trial 84 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9938068645780518, 'gamma': 0.19352476238108154}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,898] Trial 85 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.909044947880665, 'gamma': 0.08247871692776128}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,905] Trial 86 finished with value: 0.9574284676858443 and parameters: {'lambda_': 0.4897512550751938, 'gamma': 0.02369928399171662}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,912] Trial 87 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9376702608354682, 'gamma': 0.16041658085464028}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,919] Trial 88 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9289231204022531, 'gamma': 0.1662178519485552}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,926] Trial 89 finished with value: 0.9624044302256257 and parameters: {'lambda_': 0.8489243398677364, 'gamma': 0.23760188988746272}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,934] Trial 90 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8970593166535472, 'gamma': 0.19997231053107095}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,940] Trial 91 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9493997198917606, 'gamma': 0.053567112453181485}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,947] Trial 92 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9765597265892346, 'gamma': 0.05334100526264687}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,954] Trial 93 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9729388510759338, 'gamma': 0.11893970307725843}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,961] Trial 94 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9975211721101369, 'gamma': 0.13528174646218272}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,968] Trial 95 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9652696500113889, 'gamma': 0.17340245922386566}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,975] Trial 96 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9432690992569991, 'gamma': 0.11333831330845714}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,982] Trial 97 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9727973084176594, 'gamma': 0.2544706058654857}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,989] Trial 98 finished with value: 0.9639753328432574 and parameters: {'lambda_': 0.8753402487965994, 'gamma': 0.05352557894222969}. Best is trial 62 with value: 0.9696415225094471.\n", + "[I 2025-12-03 15:44:34,996] Trial 99 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.9088213150210536, 'gamma': 0.15483403519986474}. Best is trial 62 with value: 0.9696415225094471.\n", + "\n", + "============================================================\n", + "Best Trial: #62\n", + "Best CV F1 Score: 0.9696\n", + "Best Parameters:\n", + " lambda_: 0.9508\n", + " gamma: 0.0562\n", + "============================================================\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 62. Best value: 0.969642: 90%|█████████ | 90/100 [00:01<00:00, 73.00it/s]" + "\n" ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[I 2025-11-30 12:16:13,153] Trial 89 finished with value: 0.9624044302256257 and parameters: {'lambda_': 0.8489243398677364, 'gamma': 0.23760188988746272}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,164] Trial 90 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.8970593166535472, 'gamma': 0.19997231053107095}. Best is trial 62 with value: 0.9696415225094471.\n" - ] - }, + } + ], + "source": [ + "# Create study with TPE sampler (Bayesian optimization)\n", + "sampler = TPESampler(seed=42)\n", + "study = optuna.create_study(\n", + " direction='maximize',\n", + " sampler=sampler\n", + ")\n", + "\n", + "# Optimize hyperparameters\n", + "print(\"Starting Optuna optimization for RDA...\")\n", + "study.optimize(objective, n_trials=100, show_progress_bar=True)\n", + "\n", + "# Get best parameters\n", + "best_params = study.best_params\n", + "best_value = study.best_value\n", + "\n", + "print(f\"\\n{'='*60}\")\n", + "print(f\"Best Trial: #{study.best_trial.number}\")\n", + "print(f\"Best CV F1 Score: {best_value:.4f}\")\n", + "print(\"Best Parameters:\")\n", + "print(f\" lambda_: {best_params['lambda_']:.4f}\")\n", + "print(f\" gamma: {best_params['gamma']:.4f}\")\n", + "print(f\"{'='*60}\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8a0361b8", + "metadata": {}, + "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "Best trial: 62. Best value: 0.969642: 100%|██████████| 100/100 [00:01<00:00, 71.28it/s]\n", - "/Users/federicoclerici/.pyenv/versions/3.12.2/lib/python3.12/site-packages/sklearn/discriminant_analysis.py:1024: LinAlgWarning: The covariance matrix of class 1 is not full rank. Increasing the value of parameter `reg_param` might help reducing the collinearity.\n", - " warnings.warn(\n", - "\n", - "/Users/federicoclerici/.pyenv/versions/3.12.2/lib/python3.12/site-packages/sklearn/discriminant_analysis.py:1024: LinAlgWarning: The covariance matrix of class 1 is not full rank. Increasing the value of parameter `reg_param` might help reducing the collinearity.\n", + "/Users/matteo/Desktop/RegularizedDiscriminantAnalysis/.venv/lib/python3.12/site-packages/sklearn/discriminant_analysis.py:1024: LinAlgWarning: The covariance matrix of class 1 is not full rank. Increasing the value of parameter `reg_param` might help reducing the collinearity.\n", " warnings.warn(\n" ] }, @@ -283,23 +375,6 @@ "name": "stdout", "output_type": "stream", "text": [ - "[I 2025-11-30 12:16:13,209] Trial 91 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9493997198917606, 'gamma': 0.053567112453181485}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,228] Trial 92 finished with value: 0.9696415225094471 and parameters: {'lambda_': 0.9765597265892346, 'gamma': 0.05334100526264687}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,243] Trial 93 finished with value: 0.9680378848214337 and parameters: {'lambda_': 0.9729388510759338, 'gamma': 0.11893970307725843}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,255] Trial 94 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9975211721101369, 'gamma': 0.13528174646218272}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,268] Trial 95 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9652696500113889, 'gamma': 0.17340245922386566}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,280] Trial 96 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9432690992569991, 'gamma': 0.11333831330845714}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,292] Trial 97 finished with value: 0.9659405065442801 and parameters: {'lambda_': 0.9727973084176594, 'gamma': 0.2544706058654857}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,303] Trial 98 finished with value: 0.9639753328432574 and parameters: {'lambda_': 0.8753402487965994, 'gamma': 0.05352557894222969}. Best is trial 62 with value: 0.9696415225094471.\n", - "[I 2025-11-30 12:16:13,316] Trial 99 finished with value: 0.9642030548068284 and parameters: {'lambda_': 0.9088213150210536, 'gamma': 0.15483403519986474}. Best is trial 62 with value: 0.9696415225094471.\n", - "\n", - "============================================================\n", - "Best Trial: #62\n", - "Best CV F1 Score: 0.9696\n", - "Best Parameters:\n", - " lambda_: 0.9508\n", - " gamma: 0.0562\n", - "============================================================\n", "\n", "Training LDA and QDA for comparison...\n", "\n", @@ -344,14 +419,12 @@ "weighted avg 0.96 0.96 0.96 171\n", "\n", "\n", - "Comparison plot saved to 'rda_comparison.png'\n", - "\n", "Comparison plot saved to 'rda_comparison.png'\n" ] }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -361,92 +434,6 @@ } ], "source": [ - "import sys\n", - "import os\n", - "\n", - "# Add the src directory to sys.path to allow importing the package\n", - "# In Jupyter notebooks, we use the current working directory\n", - "notebook_dir = os.getcwd()\n", - "src_path = os.path.abspath(os.path.join(notebook_dir, '../src'))\n", - "if src_path not in sys.path:\n", - " sys.path.insert(0, src_path)\n", - "\n", - "from sklearn.datasets import load_breast_cancer\n", - "from sklearn.model_selection import train_test_split, cross_val_score\n", - "from sklearn.metrics import f1_score, classification_report\n", - "from sklearn.preprocessing import StandardScaler\n", - "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis, QuadraticDiscriminantAnalysis\n", - "import optuna\n", - "from optuna.trial import Trial\n", - "from optuna.samplers import TPESampler\n", - "import numpy as np\n", - "import matplotlib.pyplot as plt\n", - "from regularizeddiscriminantanalysis import RegularizedDiscriminantAnalysis\n", - "\n", - "# Load and preprocess data\n", - "X, y = load_breast_cancer(return_X_y=True)\n", - "\n", - "# Standardize features\n", - "scaler = StandardScaler()\n", - "X = scaler.fit_transform(X)\n", - "\n", - "X_train, X_test, y_train, y_test = train_test_split(\n", - " X, y, test_size=0.3, random_state=42\n", - ")\n", - "\n", - "def objective(trial: Trial) -> float:\n", - " \"\"\"Objective function for Optuna optimization.\n", - " \n", - " Parameters\n", - " ----------\n", - " trial : optuna.trial.Trial\n", - " Optuna trial object for hyperparameter suggestions.\n", - " \n", - " Returns\n", - " -------\n", - " float\n", - " Mean cross-validation accuracy (to be maximized).\n", - " \"\"\"\n", - " # Suggest hyperparameters\n", - " lambda_ = trial.suggest_float('lambda_', 0.0, 1.0)\n", - " gamma = trial.suggest_float('gamma', 0.0, 1.0)\n", - " \n", - " # Create RDA model\n", - " rda = RegularizedDiscriminantAnalysis(\n", - " lambda_=lambda_,\n", - " gamma=gamma\n", - " )\n", - " \n", - " # Cross-validate on training set\n", - " scores = cross_val_score(\n", - " rda, X_train, y_train, cv=5, scoring='f1'\n", - " )\n", - " \n", - " return scores.mean()\n", - "\n", - "# Create study with TPE sampler (Bayesian optimization)\n", - "sampler = TPESampler(seed=42)\n", - "study = optuna.create_study(\n", - " direction='maximize',\n", - " sampler=sampler\n", - ")\n", - "\n", - "# Optimize hyperparameters\n", - "print(\"Starting Optuna optimization for RDA...\")\n", - "study.optimize(objective, n_trials=100, show_progress_bar=True)\n", - "\n", - "# Get best parameters\n", - "best_params = study.best_params\n", - "best_value = study.best_value\n", - "\n", - "print(f\"\\n{'='*60}\")\n", - "print(f\"Best Trial: #{study.best_trial.number}\")\n", - "print(f\"Best CV F1 Score: {best_value:.4f}\")\n", - "print(f\"Best Parameters:\")\n", - "print(f\" lambda_: {best_params['lambda_']:.4f}\")\n", - "print(f\" gamma: {best_params['gamma']:.4f}\")\n", - "print(f\"{'='*60}\")\n", - "\n", "# Train final model with best parameters\n", "rda_best = RegularizedDiscriminantAnalysis(\n", " lambda_=best_params['lambda_'],\n", @@ -479,13 +466,13 @@ "print(f\"QDA (λ=0.0, γ=0.0): {test_f1_qda:.4f}\")\n", "print(f\"{'='*60}\")\n", "\n", - "print(f\"\\nRDA Classification Report:\")\n", + "print(\"\\nRDA Classification Report:\")\n", "print(classification_report(y_test, y_pred_rda))\n", "\n", - "print(f\"\\nLDA Classification Report:\")\n", + "print(\"\\nLDA Classification Report:\")\n", "print(classification_report(y_test, y_pred_lda))\n", "\n", - "print(f\"\\nQDA Classification Report:\")\n", + "print(\"\\nQDA Classification Report:\")\n", "print(classification_report(y_test, y_pred_qda))\n", "\n", "# Visualize optimization history and comparison\n", @@ -545,7 +532,7 @@ "ax3.grid(True, alpha=0.3, axis='y')\n", "\n", "# Add value labels on bars\n", - "for bar, acc in zip(bars, accuracies):\n", + "for bar, acc in zip(bars, accuracies, strict=False):\n", " height = bar.get_height()\n", " ax3.text(bar.get_x() + bar.get_width()/2., height,\n", " f'{acc:.4f}',\n", @@ -556,11 +543,19 @@ "print(\"\\nComparison plot saved to 'rda_comparison.png'\")\n", "plt.show()\n" ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "957b0643", + "metadata": {}, + "outputs": [], + "source": [] } ], "metadata": { "kernelspec": { - "display_name": "3.12.2", + "display_name": "RegularizedDiscriminantAnalysis", "language": "python", "name": "python3" }, @@ -574,7 +569,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.12.2" + "version": "3.12.12" } }, "nbformat": 4, diff --git a/pyproject.toml b/pyproject.toml index 71dc0be..a037f13 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -36,3 +36,29 @@ dependencies = [ [project.urls] "Homepage" = "https://github.com/F3z11/RDA_implementation" "Bug Tracker" = "https://github.com/F3z11/RDA_implementation/issues" + +[tool.ruff] +line-length = 100 +target-version = "py38" +lint.extend-select = ["E", "F", "I", "UP", "B"] +lint.ignore = ["E501"] +lint.fixable = ["ALL"] + +[tool.ruff.format] +quote-style = "double" + +[tool.ruff.lint.isort] +known-first-party = ["regularizeddiscriminantanalysis"] + +[tool.mypy] +mypy_path = "src" +strict = true +ignore_missing_imports = true +disallow_untyped_defs = true +disallow_any_unimported = true +no_implicit_optional = true +warn_unused_ignores = true +warn_return_any = true +warn_redundant_casts = true +warn_unused_configs = true +check_untyped_defs = true \ No newline at end of file diff --git a/requirements_dev.txt b/requirements_dev.txt new file mode 100644 index 0000000..900993a --- /dev/null +++ b/requirements_dev.txt @@ -0,0 +1,26 @@ +pytest==9.0.1 +ruff==0.14.7 +mypy==1.19.0 +mypy-extensions==1.1.0 +scikit-learn-stubs==0.0.3 +dataclasses==0.6 + +# install these to run the examples +contourpy==1.3.3 +cycler==0.12.1 +fonttools==4.61.0 +kiwisolver==1.4.9 +matplotlib==3.10.7 +pillow==12.0.0 +pyparsing==3.2.5 +alembic==1.17.2 +colorlog==6.10.1 +mako==1.3.10 +markupsafe==3.0.3 +optuna==4.6.0 +pyyaml==6.0.3 +sqlalchemy==2.0.44 +tqdm==4.67.1 +pandas==2.3.3 +pytz==2025.2 +tzdata==2025.2 \ No newline at end of file diff --git a/src/regularizeddiscriminantanalysis/classifier.py b/src/regularizeddiscriminantanalysis/classifier.py index 4c9dc09..37053b4 100644 --- a/src/regularizeddiscriminantanalysis/classifier.py +++ b/src/regularizeddiscriminantanalysis/classifier.py @@ -1,34 +1,40 @@ +from __future__ import annotations + import numpy as np +from scipy.linalg import pinv from sklearn.base import BaseEstimator, ClassifierMixin -from sklearn.utils.validation import check_X_y, check_array, check_is_fitted from sklearn.utils.multiclass import unique_labels -from scipy.linalg import pinv -from typing import Optional, Union, List +from sklearn.utils.validation import ( # type: ignore + check_is_fitted, + check_X_y, + validate_data, +) -class RegularizedDiscriminantAnalysis(BaseEstimator, ClassifierMixin): + +class RegularizedDiscriminantAnalysis(ClassifierMixin, BaseEstimator): """ Regularized Discriminant Analysis (RDA) - + A classifier that varies continuously between Linear Discriminant Analysis (LDA) and Quadratic Discriminant Analysis (QDA) using two regularization parameters. - + Implementation based on Friedman (1989): "Regularized Discriminant Analysis". - + Parameters ---------- lambda_ : float, default=0.0 Covariance mixing parameter (0 <= lambda_ <= 1). - 0.0: Class-specific covariances (QDA-like) - 1.0: Pooled covariance (LDA-like) - + gamma : float, default=0.0 Eigenvalue shrinkage parameter (0 <= gamma <= 1). - 0.0: No shrinkage - 1.0: Shrinks covariance towards a scalar multiple of identity matrix (Spherical) - + priors : array-like of shape (n_classes,), default=None Class prior probabilities. If None, estimated from training data. - + reg_param : float, default=1e-6 Small constant added to diagonal for numerical stability during inversion. @@ -62,17 +68,31 @@ class RegularizedDiscriminantAnalysis(BaseEstimator, ClassifierMixin): Log-determinants of regularized covariances. """ - def __init__(self, lambda_: float = 0.0, gamma: float = 0.0, - priors: Optional[np.ndarray] = None, reg_param: float = 1e-6): + def __init__( + self, + lambda_: float = 0.0, + gamma: float = 0.0, + priors: np.ndarray | None = None, + reg_param: float = 1e-6, + ): + # self.lambda_ = lambda_ this actually causes a problem cause sklearn.check_is_fitted thinks the estimator has already been fitted + self._is_fitted = False + self.lambda_ = lambda_ self.gamma = gamma self.priors = priors self.reg_param = reg_param - def fit(self, X: np.ndarray, y: np.ndarray) -> 'RegularizedDiscriminantAnalysis': + def __sklearn_is_fitted__(self) -> bool: + """ + Check fitted status and return a Boolean value. + """ + return hasattr(self, "_is_fitted") and self._is_fitted + + def fit(self, X: np.ndarray, y: np.ndarray) -> RegularizedDiscriminantAnalysis: """ Fit the RDA model according to the given training data and parameters. - + Parameters ---------- X : array-like of shape (n_samples, n_features) @@ -94,106 +114,107 @@ def fit(self, X: np.ndarray, y: np.ndarray) -> 'RegularizedDiscriminantAnalysis' # Check inputs X, y = check_X_y(X, y) self.classes_ = unique_labels(y) - self.n_classes_ = len(self.classes_) - n_samples, n_features = X.shape - + n_classes = len(self.classes_) + n_samples, self.n_features_in_ = X.shape + # 1. Compute Basic Statistics (Means, Covariances, Priors) - self.means_ = [] - self.covariances_ = [] - self.class_counts_ = [] - + self.means_ = np.empty((n_classes, self.n_features_in_), dtype=float) + self.covariances_ = np.empty( + (n_classes, self.n_features_in_, self.n_features_in_), dtype=float + ) + self._class_counts = np.empty(n_classes, dtype=int) + # We need the total scatter/covariance for pooling # Friedman uses biased (MLE) estimates in derivation, so we use ddof=0 (divide by N) # to keep math consistent with the mixing weights (N_k vs N). (coherent also with sklearn) - - for k in self.classes_: + + for idx, k in enumerate(self.classes_): X_k = X[y == k] N_k = X_k.shape[0] - - self.class_counts_.append(N_k) - self.means_.append(np.mean(X_k, axis=0)) - - # Calculate biased covariance (population covariance) + + self._class_counts[idx] = N_k + self.means_[idx] = np.mean(X_k, axis=0) + cov_k = np.cov(X_k, rowvar=False, bias=True) - - # Handle edge case of 1 sample (cov returns 0-d array or NaN if not careful) - if cov_k.ndim == 0: - cov_k = np.zeros((n_features, n_features)) - - self.covariances_.append(cov_k) - - self.means_ = np.array(self.means_) - self.covariances_ = np.array(self.covariances_) - self.class_counts_ = np.array(self.class_counts_) - + + if cov_k.ndim == 0: + cov_k = np.zeros((self.n_features_in_, self.n_features_in_)) + + self.covariances_[idx] = cov_k + # Compute Priors if self.priors is None: - self.priors_ = self.class_counts_ / n_samples + self.priors_ = self._class_counts / n_samples else: self.priors_ = np.array(self.priors) # 2. Compute Pooled Covariance (Weighted Average) # Friedman Eq (15): S = Sum(S_k) and Eq (14): Sigma_pool = S / N. # This is equivalent to weighted average of biased class covariances. - self.pooled_cov_ = np.average(self.covariances_, axis=0, weights=self.class_counts_) + self.pooled_cov_ = np.average(self.covariances_, axis=0, weights=self._class_counts) # 3. Apply Regularization (Lambda & Gamma) - Pre-compute inverses - self._apply_regularization(n_samples, n_features) - + self._apply_regularization(n_samples, self.n_features_in_) + + self._is_fitted = True + return self - def _apply_regularization(self, n_samples, n_features): + def _apply_regularization(self, n_samples: int, n_features: int) -> None: """ Computes the regularized covariance matrices, their inverses (precisions), and log-determinants. Stores them for use in prediction. """ - self.regularized_covariances_ = [] - self.precisions_ = [] - self.log_dets_ = [] + self.regularized_covariances_ = np.empty( + (len(self.classes_), n_features, n_features), + dtype=float, + ) + self.precisions_ = np.empty( + (len(self.classes_), n_features, n_features), + dtype=float, + ) + self.log_dets_ = np.empty(len(self.classes_), dtype=float) + + Identity = np.eye(n_features) - for k in range(self.n_classes_): - N_k = self.class_counts_[k] + for k in range(len(self.classes_)): + N_k = self._class_counts[k] Sigma_k = self.covariances_[k] - + # --- Step A: Lambda Mixing (Covariance vs Pool) --- # Friedman Eq (16) logic adapted for Covariances: # S_k(lambda) = (1-lambda)S_k + lambda*S # Sigma_k(lambda) = S_k(lambda) / [(1-lambda)N_k + lambda*N] - + # Let's compute the mixing weight alpha_k denom = (1 - self.lambda_) * N_k + self.lambda_ * n_samples alpha_k = ((1 - self.lambda_) * N_k) / denom beta_k = (self.lambda_ * n_samples) / denom - + Sigma_lambda_k = alpha_k * Sigma_k + beta_k * self.pooled_cov_ # --- Step B: Gamma Mixing (Shrinkage towards Identity) --- # Friedman Eq (18): (1-gamma)Sigma_k(lambda) + gamma*(tr(Sigma_k(lambda))/p)*I - + avg_eig = np.trace(Sigma_lambda_k) / n_features - Identity = np.eye(n_features) - + Sigma_final_k = (1 - self.gamma) * Sigma_lambda_k + self.gamma * avg_eig * Identity - + # Add small constant for numerical stability Sigma_final_k += self.reg_param * Identity - - self.regularized_covariances_.append(Sigma_final_k) - + + self.regularized_covariances_[k] = Sigma_final_k + # Compute Precision (Inverse) and Log-Determinant # Using pinv (pseudo-inverse) is safer for potentially singular matrices (with SVD) - self.precisions_.append(pinv(Sigma_final_k)) - + self.precisions_[k] = pinv(Sigma_final_k) + # Compute Log Determinant (using slogdet for stability) sign, logdet = np.linalg.slogdet(Sigma_final_k) if sign <= 0: # If determinant is 0 or negative (shouldn't happen with reg_param), fallback logdet = np.log(max(np.linalg.det(Sigma_final_k), 1e-10)) - self.log_dets_.append(logdet) - - self.regularized_covariances_ = np.array(self.regularized_covariances_) - self.precisions_ = np.array(self.precisions_) - self.log_dets_ = np.array(self.log_dets_) + self.log_dets_[k] = logdet def predict(self, X: np.ndarray) -> np.ndarray: """ @@ -210,10 +231,19 @@ def predict(self, X: np.ndarray) -> np.ndarray: Predicted class labels. """ check_is_fitted(self) - X = check_array(X) - + # X = check_array(X) + X = validate_data(self, X, reset=False) + scores = self._decision_function(X) - return self.classes_[np.argmax(scores, axis=1)] + try: + res: np.ndarray = self.classes_[np.argmax(scores, axis=1)] + except IndexError as err: + raise IndexError( + "The number of classes in the training data does not match " + "the number of classes during prediction." + ) from err + + return res def _decision_function(self, X: np.ndarray) -> np.ndarray: """ @@ -221,28 +251,28 @@ def _decision_function(self, X: np.ndarray) -> np.ndarray: Score_k = - (log|Sigma_k| + (x-mu_k)^T Sigma_k^-1 (x-mu_k)) + 2 * log(prior_k) """ n_samples = X.shape[0] - scores = np.zeros((n_samples, self.n_classes_)) - - for k in range(self.n_classes_): + scores = np.zeros((n_samples, len(self.classes_))) + + for k in range(len(self.classes_)): # Get pre-computed parameters precision = self.precisions_[k] log_det = self.log_dets_[k] mean = self.means_[k] prior = self.priors_[k] - + # Centered data X_centered = X - mean - + # Mahalanobis Distance squared: (x-mu)^T * Sigma^-1 * (x-mu) # Efficient calculation: sum( (X_centered @ precision) * X_centered, axis=1) mahalanobis = np.sum(np.dot(X_centered, precision) * X_centered, axis=1) - + # Score (Quadratic term + Log Det term + Prior term) # We omit constant terms (like n_features * log(2pi)) as they don't affect argmax - scores[:, k] = - (log_det + mahalanobis) + 2 * np.log(prior) - + scores[:, k] = -(log_det + mahalanobis) + 2 * np.log(prior) + return scores - + def predict_proba(self, X: np.ndarray) -> np.ndarray: """ Return probability estimates for the test data X. @@ -259,17 +289,18 @@ def predict_proba(self, X: np.ndarray) -> np.ndarray: where classes are ordered as they are in self.classes_. """ check_is_fitted(self) - X = check_array(X) - + # X = check_array(X) + X = validate_data(self, X, reset=False) + # Get log-posterior scores (unnormalized) decision_scores = self._decision_function(X) - + # Apply Softmax (Numerically Stable) # Subtract max per row to avoid overflow in exp() max_scores = np.max(decision_scores, axis=1, keepdims=True) exp_scores = np.exp(decision_scores - max_scores) - probs = exp_scores / np.sum(exp_scores, axis=1, keepdims=True) - + probs: np.ndarray = exp_scores / np.sum(exp_scores, axis=1, keepdims=True) + return probs def predict_log_proba(self, X: np.ndarray) -> np.ndarray: @@ -286,4 +317,5 @@ def predict_log_proba(self, X: np.ndarray) -> np.ndarray: log_probs : ndarray of shape (n_samples, n_classes) Returns the log-probability of the sample for each class in the model. """ - return np.log(self.predict_proba(X)) + log_proba: np.ndarray = np.log(self.predict_proba(X)) + return log_proba diff --git a/tests/test_rda.py b/tests/test_rda.py index 3315f72..d6ff4ae 100644 --- a/tests/test_rda.py +++ b/tests/test_rda.py @@ -1,130 +1,121 @@ +import os +import sys +import warnings + import numpy as np -from sklearn.discriminant_analysis import LinearDiscriminantAnalysis, QuadraticDiscriminantAnalysis +import pytest from sklearn.datasets import make_classification +from sklearn.discriminant_analysis import LinearDiscriminantAnalysis, QuadraticDiscriminantAnalysis from sklearn.model_selection import train_test_split -from sklearn.metrics import accuracy_score -import warnings - -import sys -import os # Add the src directory to sys.path to allow importing the package -sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '../src'))) +sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "../src"))) -# Import your class try: from regularizeddiscriminantanalysis import RegularizedDiscriminantAnalysis except ImportError: - raise ImportError("Could not import 'RegularizedDiscriminantAnalysis' from 'regularizeddiscriminantanalysis'. Make sure the package is installed or in the python path.") + raise ImportError( + "Could not import 'RegularizedDiscriminantAnalysis'. Ensure it is installed or discoverable." + ) from None -def test_rda_implementation(): - print("="*60) - print("RDA IMPLEMENTATION TEST SUITE") - print("="*60) - # 1. Generate Synthetic Data - # -------------------------- +@pytest.fixture(scope="module") +def sample_data() -> tuple[np.ndarray, np.ndarray, np.ndarray, np.ndarray]: + """Generate a small multiclass dataset for reproducible tests.""" X, y = make_classification( - n_samples=200, n_features=5, n_informative=3, n_redundant=0, - n_classes=3, random_state=42, class_sep=2.0 + n_samples=200, + n_features=5, + n_informative=3, + n_redundant=0, + n_classes=3, + random_state=42, + class_sep=2.0, ) X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42) - - print(f"Data generated: {len(y_train)} training samples, 3 classes, 5 features.\n") - - # 2. Test QDA Equivalence (lambda=0, gamma=0) - # ------------------------------------------- - print(f"TEST 1: QDA Equivalence (lambda=0, gamma=0)") - - # Sklearn QDA + + return X_train, X_test, y_train, y_test + + +def test_rda_match(sample_data: tuple[np.ndarray, np.ndarray, np.ndarray, np.ndarray]) -> None: + X_train, X_test, y_train, y_test = sample_data + qda = QuadraticDiscriminantAnalysis(store_covariance=True) qda.fit(X_train, y_train) qda_pred = qda.predict(X_test) - - # Your RDA + rda_qda = RegularizedDiscriminantAnalysis(lambda_=0.0, gamma=0.0) rda_qda.fit(X_train, y_train) rda_pred = rda_qda.predict(X_test) - + match_count = np.sum(qda_pred == rda_pred) total = len(y_test) - print(f" - Sklearn QDA Accuracy: {accuracy_score(y_test, qda_pred):.4f}") - print(f" - Your RDA Accuracy: {accuracy_score(y_test, rda_pred):.4f}") - print(f" - Prediction Match: {match_count}/{total} ({match_count/total:.1%})") - - if match_count == total: - print(" ✅ SUCCESS: Matches Sklearn QDA exactly.") - else: - print(" ⚠️ WARNING: Does not match exactly. (Small differences are expected due to biased/unbiased estimator choices, but accuracy should be similar).") - print("-" * 60) + assert match_count == total, ( + "RDA with lambda=0, gamma=0 should reproduce QDA predictions exactly." + ) + + +def test_lda_equivalence( + sample_data: tuple[np.ndarray, np.ndarray, np.ndarray, np.ndarray], +) -> None: + X_train, X_test, y_train, y_test = sample_data - # 3. Test LDA Equivalence (lambda=1, gamma=0) - # ------------------------------------------- - print(f"TEST 2: LDA Equivalence (lambda=1, gamma=0)") - - # Sklearn LDA lda = LinearDiscriminantAnalysis(store_covariance=True) lda.fit(X_train, y_train) lda_pred = lda.predict(X_test) - - # Your RDA + rda_lda = RegularizedDiscriminantAnalysis(lambda_=1.0, gamma=0.0) rda_lda.fit(X_train, y_train) - rda_pred_lda = rda_lda.predict(X_test) - - match_count = np.sum(lda_pred == rda_pred_lda) - print(f" - Sklearn LDA Accuracy: {accuracy_score(y_test, lda_pred):.4f}") - print(f" - Your RDA Accuracy: {accuracy_score(y_test, rda_pred_lda):.4f}") - print(f" - Prediction Match: {match_count}/{total} ({match_count/total:.1%})") - - if match_count == total: - print(" ✅ SUCCESS: Matches Sklearn LDA exactly.") - else: - print(" ⚠️ WARNING: Mismatch found. Check your 'pooled covariance' calculation. Sklearn uses weighted average.") + rda_pred = rda_lda.predict(X_test) - print("-" * 60) + match_count = np.sum(lda_pred == rda_pred) + total = len(y_test) + + # Logging information for debugging (pytest captures this) + print(f"Sklearn LDA vs RDA(lambda=1): matched predictions {match_count}/{total}") + + assert match_count == total, ( + "RDA with lambda=1, gamma=0 should reproduce LDA predictions exactly." + ) - # 4. Math Logic & Scale Bug Check - # ------------------------------- - print(f"TEST 3: Scale/Math Logic Check") - - # We create a tiny dummy dataset where we know the answer - X_dummy = np.array([[1., 2.], [2., 3.]]) # Class 0 + +def test_scale_logic() -> None: + """Ensure covariance scaling is correct on a toy example.""" + X_dummy = np.array([[1.0, 2.0], [2.0, 3.0]]) y_dummy = np.array([0, 0]) - - # Fit RDA + rda_debug = RegularizedDiscriminantAnalysis(lambda_=0.0, gamma=0.0) rda_debug.fit(X_dummy, y_dummy) - - # Check the stored covariance for class 0 - # Standard Covariance of [[1,2], [2,3]] is [[0.5, 0.5], [0.5, 0.5]] (Using N=2, biased? or N-1=1 unbiased?) - # Sklearn uses Unbiased by default usually, or Biased depending on solver. - # Numpy cov uses (N-1) by default. - - try: - # Accessing internal attribute - adjust name if you changed it (e.g. regularized_covariances_ or covariances_) - if hasattr(rda_debug, 'regularized_covariances_'): - stored_cov = rda_debug.regularized_covariances_[0] - elif hasattr(rda_debug, 'covariances_'): - stored_cov = rda_debug.covariances_[0] - else: - print(" ❓ SKIPPING: Could not find 'covariances_' attribute to inspect.") - return - - print(f" - Input Data (Class 0):\n{X_dummy}") - print(f" - Your Computed Covariance:\n{stored_cov}") - - # Heuristic check: Elements should be around 0.5, not 0.005 - if np.all(np.abs(stored_cov) < 0.01): - print(" ❌ CRITICAL FAIL: Covariance values are tiny. You likely have the 'Scale Bug' (dividing by weight twice).") - elif np.all(stored_cov == 0): - print(" ❌ FAIL: Covariance is zero.") - else: - print(" ✅ PASS: Covariance scale looks reasonable.") - - except Exception as e: - print(f" ❌ ERROR during inspection: {e}") - -if __name__ == "__main__": - test_rda_implementation() \ No newline at end of file + + if hasattr(rda_debug, "regularized_covariances_"): + stored_cov = rda_debug.regularized_covariances_[0] + elif hasattr(rda_debug, "covariances_"): + stored_cov = rda_debug.covariances_[0] + else: + warnings.warn( + "Skipping covariance scale test: RDA exposes neither " + "'regularized_covariances_' nor 'covariances_'.", + stacklevel=2, + ) + return + + # Log the values for pytest output (not an assertion condition) + print("Input data:\n", X_dummy) + print("Computed covariance matrix:\n", stored_cov) + + # Core assertion: covariance must have reasonable magnitude + if np.all(stored_cov == 0): + raise AssertionError("Covariance matrix contains only zeros.") + + if np.all(np.abs(stored_cov) < 0.01): + raise AssertionError( + "Covariance values are extremely small. " + "This suggests a scaling issue (possibly dividing by a weight twice)." + ) + + # If covariance looks valid but shape is off, warn + if stored_cov.shape != (2, 2): + warnings.warn( + f"Unexpected covariance matrix shape: {stored_cov.shape}.", + stacklevel=1, + ) diff --git a/tests/test_sklearn_compatibility.py b/tests/test_sklearn_compatibility.py new file mode 100644 index 0000000..c745e4c --- /dev/null +++ b/tests/test_sklearn_compatibility.py @@ -0,0 +1,170 @@ +# type: ignore + +import os +import pickle +import sys +from dataclasses import dataclass +from typing import Any + +import numpy as np +import pytest +from sklearn.base import clone +from sklearn.datasets import make_classification +from sklearn.model_selection import GridSearchCV, StratifiedKFold +from sklearn.pipeline import Pipeline +from sklearn.preprocessing import StandardScaler +from sklearn.utils.estimator_checks import check_estimator, check_is_fitted, parametrize_with_checks + +# Add the src directory to sys.path to allow importing the package +sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "../src"))) + +try: + from regularizeddiscriminantanalysis import RegularizedDiscriminantAnalysis +except ImportError: + raise ImportError( + "Could not import 'RegularizedDiscriminantAnalysis'. Ensure it is installed or discoverable." + ) from None + + +@pytest.mark.parametrize( + "estimator_cls", + [RegularizedDiscriminantAnalysis], +) +def test_clone_rda(estimator_cls: type[Any]) -> None: + estimator = estimator_cls() + auto = clone(estimator) + assert isinstance(auto, RegularizedDiscriminantAnalysis) + + +@pytest.mark.parametrize( + "estimator_cls", + [RegularizedDiscriminantAnalysis], +) +def test_check_estimator_basic(estimator_cls: type[Any]) -> None: + estimator = estimator_cls() + check_estimator(estimator) + + +@parametrize_with_checks([RegularizedDiscriminantAnalysis()]) +def test_sklearn_compatibility(estimator: Any, check: Any) -> None: + check(estimator) + + +@pytest.mark.parametrize( + "estimator_cls", + [RegularizedDiscriminantAnalysis], +) +def test_pipeline(estimator_cls: type[Any]) -> None: + est = estimator_cls() + + pipe = Pipeline([("scaler", StandardScaler()), ("clf", est)]) + pipe2 = clone(pipe) + + params = pipe2.get_params(deep=True) + assert "clf__lambda_" in params + assert "clf__gamma" in params + assert "clf__reg_param" in params + assert "scaler__with_mean" in params + assert "scaler__with_std" in params + + +@dataclass(frozen=True) +class GridSearchCheckResult: + best_params: dict[str, Any] + best_score: float + + +def assert_gridsearch_compatible( + estimator, + X: np.ndarray, + y: np.ndarray, + param_grid: dict[str, list[Any]], + scoring: str = "accuracy", + cv_splits: int = 5, + refit: bool = True, +) -> GridSearchCheckResult: + try: + est2 = clone(estimator) + except Exception as e: + raise AssertionError( + "Estimator is not cloneable. Ensure __init__ only assigns parameters " + "verbatim and you don't override __getstate__/__setstate__ oddly." + ) from e + + params = est2.get_params(deep=True) + missing = [k for k in param_grid.keys() if k not in params] + if missing: + raise AssertionError( + f"param_grid contains keys not in estimator.get_params(): {missing}.\n" + f"Available params: {sorted(params.keys())}" + ) + + first_setting = {k: v[0] for k, v in param_grid.items()} + try: + est2.set_params(**first_setting) + except Exception as e: + raise AssertionError("Estimator.set_params failed for provided grid keys.") from e + + cv = StratifiedKFold(n_splits=cv_splits, shuffle=True, random_state=0) + gs = GridSearchCV( + estimator=estimator, + param_grid=param_grid, + scoring=scoring, + cv=cv, + refit=refit, + error_score="raise", + n_jobs=1, + return_train_score=True, + ) + gs.fit(X, y) + + if not hasattr(gs, "best_estimator_"): + raise AssertionError("GridSearchCV did not produce best_estimator_.") + + try: + check_is_fitted(gs.best_estimator_) + except Exception as e: + raise AssertionError( + "best_estimator_ is not recognized as fitted. " + "Ensure fit() sets fitted attributes and/or __sklearn_is_fitted__." + ) from e + + if not hasattr(gs, "best_params_") or not hasattr(gs, "best_score_"): + raise AssertionError("GridSearchCV did not populate best_params_ / best_score_.") + + return GridSearchCheckResult( + best_params=dict(gs.best_params_), best_score=float(gs.best_score_) + ) + + +@pytest.mark.parametrize("estimator_cls", [RegularizedDiscriminantAnalysis]) +def test_gridsearch(estimator_cls: type[Any]) -> None: + X, y = make_classification( + n_samples=400, n_features=10, n_informative=6, n_redundant=0, n_classes=3, random_state=0 + ) + + est = estimator_cls() + grid = { + "lambda_": [0.0, 0.5, 1.0], + "gamma": [0.0, 0.2, 0.8], + "reg_param": [1e-8, 1e-6], + } + + res = assert_gridsearch_compatible(est, X, y, param_grid=grid) + assert set(res.best_params.keys()) == set(grid.keys()) + + +@pytest.mark.parametrize( + "estimator_cls", + [RegularizedDiscriminantAnalysis], +) +def test_clone_and_pickle(estimator_cls: type[Any]) -> None: + rng = np.random.RandomState(2) + X = rng.randn(80, 8) + y = np.ones(80) + est = estimator_cls() + est.fit(X, y) + s = pickle.dumps(est) + est2 = pickle.loads(s) + + assert isinstance(est2, estimator_cls)