micromlkit 0.1.1

Mini machine learning library

microMlKit

microMlKit is a lightweight, educational machine learning toolkit built with NumPy and a scikit-learn-inspired API.

Package name for imports and installation metadata: micromlkit (all lowercase).

What this project is

This library is intended for learning, experimentation, and understanding ML internals—not production workloads.

It includes:

• Estimator and mixin base classes
• Linear models
• Clustering, tree, ensemble, neighbors, SVM, and decomposition modules
• Preprocessing helpers
• Classification and regression metrics
• Model selection helpers (train_test_split, KFold, StratifiedKFold, cross-validation, grid search)

Installation

Install from PyPI:

pip install micromlkit==0.1.0

Quick start

Linear regression example

import numpy as np

from micromlkit.linear_model.linear_regression import LinearRegression
from micromlkit.metrics.regression import mean_squared_error, r2_score
from micromlkit.model_selection.split import train_test_split

X = np.array([[1.0], [2.0], [3.0], [4.0], [5.0]])
y = np.array([2.0, 4.1, 6.1, 8.2, 10.1])

X_train, X_test, y_train, y_test = train_test_split(
	X, y, test_size=0.4, random_state=42
)

model = LinearRegression().fit(X_train, y_train)
predictions = model.predict(X_test)

print("MSE:", mean_squared_error(y_test, predictions))
print("R2:", r2_score(y_test, predictions))

Base pipeline example

BasePipeline (from micromlkit.base) supports a simple transform-then-estimate flow.

import numpy as np

from micromlkit.base import BasePipeline


class DoubleTransformer:
	def fit_transform(self, X, y=None):
		return np.asarray(X) * 2

	def transform(self, X):
		return np.asarray(X) * 2


class FirstColumnModel:
	def fit(self, X, y=None):
		return self

	def predict(self, X):
		X = np.asarray(X)
		return X[:, 0]


pipeline = BasePipeline(
	steps=[("double", DoubleTransformer()), ("model", FirstColumnModel())]
)

X = np.array([[1, 2], [3, 4]])
y = np.array([10, 20])

pipeline.fit(X, y)
print(pipeline.predict(X))

Module highlights

• micromlkit.base: BaseEstimator, BaseModel, BaseTransformer, mixins, BasePipeline
• micromlkit.linear_model: LinearRegression, Lasso, Ridge, LogisticRegression
• micromlkit.metrics: classification + regression metrics
• micromlkit.model_selection: splitters, cross-validation, search utilities
• micromlkit.preprocessing: encoder, imputer, scaler helpers
• micromlkit.cluster, tree, ensemble, neighbors, svm, decomposition

Testing

Run the test suite:

pytest

Current status

• API is intentionally minimal and may evolve.
• Some advanced conveniences (for example, richer nested-parameter handling and expanded pipeline ergonomics) are still limited.

Implementation checklist

Status below is based on current source files in micromlkit/.

Implemented (non-empty)

• micromlkit/__init__.py
• micromlkit/base.py
• micromlkit/linear_model/linear_regression.py
• micromlkit/linear_model/__init__.py
• micromlkit/metrics/classification.py
• micromlkit/metrics/regression.py
• micromlkit/metrics/__init__.py
• micromlkit/model_selection/split.py
• micromlkit/model_selection/cross_validation.py
• micromlkit/model_selection/search.py

Pending implementation (empty files)

Cluster

• micromlkit/cluster/agglomerative.py
• micromlkit/cluster/dbscan.py
• micromlkit/cluster/kmeans.py
• micromlkit/cluster/__init__.py (exports)

Decomposition

• micromlkit/decomposition/pca.py
• micromlkit/decomposition/__init__.py (exports)

Ensemble

• micromlkit/ensemble/random_forest.py
• micromlkit/ensemble/gradient_boosting.py
• micromlkit/ensemble/__init__.py (exports)

Linear model

• micromlkit/linear_model/lasso.py
• micromlkit/linear_model/ridge.py
• micromlkit/linear_model/logistic_regression.py

Neighbors

• micromlkit/neighbors/knn_classifier.py
• micromlkit/neighbors/knn_regressor.py
• micromlkit/neighbors/__init__.py (exports)

Pipeline

• micromlkit/pipeline/pipeline.py
• micromlkit/pipeline/__init__.py (exports)

Preprocessing

• micromlkit/preprocessing/encoder.py
• micromlkit/preprocessing/imputer.py
• micromlkit/preprocessing/scaler.py
• micromlkit/preprocessing/__init__.py (exports)

SVM

• micromlkit/svm/svm.py
• micromlkit/svm/__init__.py (exports)

Tree

• micromlkit/tree/decision_tree_classifier.py
• micromlkit/tree/decision_tree_regressor.py
• micromlkit/tree/__init__.py (exports)

Utils

• micromlkit/utils/math.py
• micromlkit/utils/validation.py
• micromlkit/utils/__init__.py (exports)

Model selection package init

• micromlkit/model_selection/__init__.py (exports)

Project layout

micromlkit/
├── base.py
├── cluster/
├── decomposition/
├── ensemble/
├── linear_model/
├── metrics/
├── model_selection/
├── neighbors/
├── pipeline/
├── preprocessing/
├── svm/
└── tree/