scratchkit 0.2.0__tar.gz

scratchkit-0.2.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

scratchkit-0.2.0/PKG-INFO

@@ -0,0 +1,241 @@
+Metadata-Version: 2.4
+Name: scratchkit
+Version: 0.2.0
+Summary: Pure-NumPy from-scratch implementations of ML/AI/RL/Bayesian algorithms — no PyTorch, no TensorFlow, no scikit-learn. (import name: mlscratch)
+Author-email: Mattral <mattral@example.com>
+License: Apache-2.0
+Project-URL: Homepage, https://github.com/Mattral/ML-AI-Algorithms-from-scratch
+Project-URL: Documentation, https://mattral.github.io/ML-AI-Algorithms-from-scratch/
+Project-URL: Repository, https://github.com/Mattral/ML-AI-Algorithms-from-scratch
+Project-URL: Issues, https://github.com/Mattral/ML-AI-Algorithms-from-scratch/issues
+Project-URL: Changelog, https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/CHANGELOG.md
+Keywords: machine-learning,deep-learning,reinforcement-learning,bayesian,numpy,from-scratch,education,algorithms,neural-network,unsupervised,supervised
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+Classifier: Intended Audience :: Education
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Education
+Classifier: Typing :: Typed
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy>=1.23
+Provides-Extra: dev
+Requires-Dist: pytest>=7; extra == "dev"
+Requires-Dist: pytest-cov>=4; extra == "dev"
+Requires-Dist: pytest-benchmark>=4; extra == "dev"
+Requires-Dist: hypothesis>=6; extra == "dev"
+Requires-Dist: scikit-learn>=1.3; extra == "dev"
+Requires-Dist: ruff>=0.4; extra == "dev"
+Requires-Dist: black>=24; extra == "dev"
+Requires-Dist: mypy>=1.8; extra == "dev"
+Requires-Dist: build>=1; extra == "dev"
+Requires-Dist: twine>=5; extra == "dev"
+Provides-Extra: docs
+Requires-Dist: mkdocs>=1.6; extra == "docs"
+Requires-Dist: mkdocs-material>=9.5; extra == "docs"
+Requires-Dist: mkdocstrings[python]>=0.25; extra == "docs"
+Provides-Extra: notebooks
+Requires-Dist: jupyter>=1.0; extra == "notebooks"
+Requires-Dist: matplotlib>=3.7; extra == "notebooks"
+Requires-Dist: pandas>=2.0; extra == "notebooks"
+Provides-Extra: all
+Requires-Dist: scratchkit[dev,docs,notebooks]; extra == "all"
+Dynamic: license-file
+
+# ML-AI-Algorithms-from-scratch
+
+**60+ ML/AI/DL/RL/Bayesian algorithms implemented from scratch in NumPy — plus `mlscratch`, a pip-installable package (`pip install scratchkit`) with a consistent, scikit-learn-style API and 1,100+ tests.**
+
+[![CI](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/actions/workflows/ci.yml/badge.svg)](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/actions)
+[![PyPI](https://img.shields.io/pypi/v/scratchkit.svg)](https://pypi.org/project/scratchkit/)
+[![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/LICENSE)
+[![Python 3.10+](https://img.shields.io/badge/Python-3.10+-blue.svg)](https://python.org)
+[![Stars](https://img.shields.io/github/stars/Mattral/ML-AI-Algorithms-from-scratch?style=social)](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/stargazers)
+
+> **What's here:** readable, standalone implementations of algorithms you already know by name, written to show the math in code, not to be fast.
+>
+> **What's new:** `src/mlscratch/` — a pip-installable package with `fit()`/`predict()`/`transform()` APIs, full type hints, and a test suite that cross-checks correctness against scikit-learn wherever a reference implementation exists.
+
+---
+
+## What makes this different from the dozens of similar repos
+
+There are many "ML from scratch" repos on GitHub. The honest differentiators here:
+
+- **Bayesian methods are first-class.** Most from-scratch repos stop at supervised learning + neural nets. This one includes Bayesian Neural Networks, Gaussian Processes, Hidden Markov Models, Bayesian Networks, and Kalman Filters — algorithms most tutorials skip because they're harder to implement correctly.
+- **RL goes beyond DQN.** DDPG, TD3, SAC, and PPO are included alongside tabular Q-Learning and DQN — non-trivial to implement correctly from scratch, and rare to see done well in a single repo.
+- **The `src/mlscratch` package is real, not a wrapper.** Every estimator is implemented in pure NumPy — no calling out to scikit-learn at runtime. scikit-learn only appears in the *test suite*, as a correctness oracle, never as a dependency of the library itself.
+- **Kernel SVM via real SMO, gradient boosting with proper Newton-step leaves, multiclass-native AdaBoost (SAMME.R)** — the ensemble/kernel methods aren't toy simplifications; several are verified to match scikit-learn's output to floating-point tolerance on real benchmarks.
+
+---
+
+## Quick start
+
+### Browse the standalone scripts (no install needed)
+
+```bash
+git clone https://github.com/Mattral/ML-AI-Algorithms-from-scratch
+cd ML-AI-Algorithms-from-scratch
+
+pip install numpy matplotlib scikit-learn   # only deps, for the standalone scripts
+
+python "Supervised/LinearRegression/linear_regression.py"
+python "Neural Networks/Transformer/transformer.py"
+python "Reinforcement/PPO/ppo.py"
+```
+
+### Use the package
+
+```bash
+pip install scratchkit                # from PyPI — the import name is still `mlscratch`
+```
+
+```bash
+# — or, for local development —
+pip install -e .                  # installs src/mlscratch in editable mode
+# pip install -e ".[dev]"         # + pytest, ruff, black, mypy, for development
+
+pytest tests/ -v                  # run the test suite
+python -m mlscratch info          # package + sub-package summary
+python -m mlscratch list supervised
+```
+
+```python
+from mlscratch.supervised import RandomForestClassifier
+from mlscratch.preprocessing import StandardScaler, train_test_split
+from mlscratch.metrics import classification_report
+
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, stratify=y)
+
+scaler = StandardScaler().fit(X_train)
+model = RandomForestClassifier(n_estimators=200, max_depth=6, oob_score=True)
+model.fit(scaler.transform(X_train), y_train)
+
+print(f"OOB score: {model.oob_score_:.3f}")
+print(classification_report(y_test, model.predict(scaler.transform(X_test))))
+```
+
+See [`examples/`](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/tree/main/examples) for six runnable end-to-end scripts covering decision trees, random forests, kernel SVMs, gradient boosting, AdaBoost, and a full no-sklearn classification + regression pipeline.
+
+---
+
+## What's implemented
+
+### `mlscratch` package (`src/mlscratch/`)
+
+| Sub-package | Contents | Tests |
+|---|---|---|
+| `mlscratch.supervised` | Linear/Ridge/Lasso/ElasticNet/Logistic regression, KNN, **DecisionTree** (classifier + regressor), **RandomForest** (bagging + OOB scoring), kernel **SVC** (SMO; linear/poly/rbf/sigmoid, one-vs-rest multiclass), **GradientBoosting** (classifier + regressor, squared/absolute-error loss), **AdaBoost** (SAMME / SAMME.R, multiclass-native) | 162 |
+| `mlscratch.unsupervised` | K-Means++, K-Medoids, DBSCAN, Agglomerative Clustering, PCA, t-SNE, FastICA, Gaussian Mixture Model (EM), Apriori | 120 |
+| `mlscratch.bayesian` | Naive Bayes (Gaussian/Multinomial/Bernoulli), Bayesian Linear Regression, Bayesian Network, Bayesian Neural Network (mean-field VI), Gaussian Process Regression, Hidden Markov Model, Kalman Filter | 171 |
+| `mlscratch.reinforcement` | Q-Learning, Double Q-Learning, DQN (Double + Dueling + PER), DDPG, TD3, PPO (GAE-λ), SAC, plus shared `GridWorld`/`ReplayBuffer`/`PrioritizedReplayBuffer` utilities | 218 |
+| `mlscratch.neural` | Single/Multi-Layer Perceptron, Autoencoder (vanilla/denoising/variational), RNN/LSTM/Encoder-Decoder, a small CNN (Conv2D/Pool/BatchNorm), Attention + Transformer encoder, GAN, Hopfield Network, Restricted Boltzmann Machine, RBF Network, Complex-Valued NN | 372 |
+| `mlscratch.metrics` | accuracy/precision/recall/F1, confusion matrix, `classification_report`, ROC/AUC, log loss, MSE/RMSE/MAE/MAPE, R², explained variance — every metric checked against scikit-learn | 48 |
+| `mlscratch.preprocessing` | StandardScaler, MinMaxScaler, RobustScaler, Normalizer, LabelEncoder, OneHotEncoder, PolynomialFeatures, `train_test_split` (with stratification) | 62 |
+
+**1,153 tests total.** A handful (~18) fail under the newest NumPy/SciPy releases in this environment due to upstream API drift in unrelated modules (Bayesian networks, reinforcement learning buffers, ICA) — tracked as known issues, not part of this release's scope.
+
+### Standalone scripts (original, by category)
+
+These are the original from-scratch scripts the package above was distilled from — browse them like a reference, run them directly, no install required.
+
+- **`Supervised/`** — Linear/Ridge/Lasso Regression, Logistic Regression, k-NN, Decision Trees, Random Forest, Naive Bayes, SVM
+- **`Unsupervised/`** — K-Means++, K-Medoids, DBSCAN, Hierarchical Clustering, PCA, t-SNE, ICA, Gaussian Mixture Model, EM, Self-Organising Map, Apriori
+- **`Neural Networks/`** — Single/Multi-Layer Perceptron, Simple RNN, LSTM, Simple CNN, Encoder-Decoder, Self-Attention, Transformer, Autoencoder, GAN, Boltzmann Machine, Hopfield Network, RBF Networks
+- **`Reinforcement/`** — Q-Learning, DQN, DDPG, PPO, SAC
+- **`Bayesian Learning/`** — Bayesian Inference, Bayesian Linear Regression, Bayesian Network, Bayesian Neural Networks, Gibbs Sampling, Metropolis-Hastings, Variational Inference
+
+---
+
+## Design philosophy
+
+Every implementation applies the same principles:
+
+- Explicit loops over vectorised one-liners when clarity improves
+- Model logic, loss computation, and parameter updates in separate functions
+- The package layer (`src/mlscratch`) calls **only** NumPy at runtime — scikit-learn appears solely in the test suite, as a correctness oracle
+- Short files: most standalone scripts are 100–300 lines; package modules favor one well-documented class per concern
+
+**This trades raw performance for readability and correctness-by-inspection. That's intentional.**
+
+If you're looking for production-speed implementations, use scikit-learn, PyTorch, or JAX. If you want to read the math in code form — or verify it against a reference implementation in the test suite — this is the repo.
+
+---
+
+## Recommended learning path
+
+If you're working through this systematically:
+
+1. Start with `Supervised/LinearRegression` (or `mlscratch.supervised.LinearRegression`) — the simplest possible end-to-end example
+2. Move to `LogisticRegression` — same structure, adds sigmoid + cross-entropy
+3. Then `DecisionTreeClassifier` → `RandomForestClassifier` → `GradientBoostingClassifier`/`AdaBoostClassifier` — the tree-ensemble family, building on a shared CART implementation
+4. Then `Neural Networks/SingleLayerPerceptron` → `MultiLayerPerceptron` — backprop from first principles
+5. Then any of: Unsupervised (PCA → GMM → t-SNE), Reinforcement (Q-Learning → DQN → PPO/SAC), or Bayesian (Naive Bayes → Bayesian Linear Regression → Variational Inference)
+
+Each folder/module is reasonably self-contained — jump to any algorithm without reading the others first.
+
+---
+
+## Repository layout
+
+```
+ML-AI-Algorithms-from-scratch/
+│
+├── Supervised/              Standalone scripts: LinearRegression, SVM, etc.
+├── Unsupervised/            Standalone scripts: KMeans++, DBSCAN, t-SNE, etc.
+├── Neural Networks/         Standalone scripts: MLP, LSTM, Transformer, GAN, etc.
+├── Reinforcement/           Standalone scripts: DQN, DDPG, PPO, SAC, etc.
+├── Bayesian Learning/       Standalone scripts: BNN, VI, MCMC, etc.
+│
+├── src/mlscratch/           Pip-installable package
+│   ├── supervised/          Linear models, KNN, trees, ensembles, kernel SVM
+│   ├── unsupervised/        Clustering, dimensionality reduction, association rules
+│   ├── bayesian/            Naive Bayes, BLR, BNN, GP, HMM, Bayesian Networks, Kalman
+│   ├── reinforcement/       Q-Learning, DQN, DDPG, TD3, PPO, SAC
+│   ├── neural/              Perceptrons, autoencoders, RNN/CNN, attention, GAN, ...
+│   ├── metrics/             Classification & regression evaluation metrics
+│   └── preprocessing/       Scalers, encoders, polynomial features, train_test_split
+│
+├── examples/                Runnable end-to-end scripts (no sklearn at runtime)
+├── tests/                   1,153 tests, mirroring the src/mlscratch layout
+├── docs/                    Roadmap (MkDocs site planned, see roadmap.md)
+├── pyproject.toml           Package metadata + deps
+├── CHANGELOG.md             Keep-a-Changelog formatted release history
+├── roadmap.md               P0 / P1 / P2 backlog
+├── .github/workflows/       CI: lint → test matrix → build → PyPI release
+└── README.md
+```
+
+---
+
+## Contributing
+
+The most useful contributions right now:
+
+- **Add a standalone script** for an algorithm not yet covered (check the folder first)
+- **Port a standalone script** into `src/mlscratch` with a matching test file in `tests/`
+- **Fix a numerical issue** — some implementations have known edge cases under newer NumPy/SciPy releases (see the known-issues note above; open an issue or PR)
+
+Standard flow: fork → branch → PR. CI runs `ruff`, `black --check`, and the full `pytest` suite on every PR. See [`CONTRIBUTING.md`](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/CONTRIBUTING.md) for the full guide, and [`roadmap.md`](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/roadmap.md) for what's planned next.
+
+---
+
+## Honest scope
+
+The standalone scripts under `Supervised/`, `Neural Networks/`, etc. are a **learning reference**, not a performance library: some use toy datasets, a few have hardcoded hyperparameters to keep the code short, and none are tuned for speed at scale.
+
+The `src/mlscratch` package is more rigorous (typed, tested, cross-checked against scikit-learn) but is still pure-Python/NumPy — it will not outrun scikit-learn or XGBoost on large datasets, and that was never the goal. The public API is stabilising but may still change between minor versions before a 1.0 release; pin a version if you're building on top of it.
+
+---
+
+## License
+
+Apache 2.0 — see [LICENSE](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/LICENSE).

scratchkit-0.2.0/README.md

@@ -0,0 +1,189 @@
+# ML-AI-Algorithms-from-scratch
+
+**60+ ML/AI/DL/RL/Bayesian algorithms implemented from scratch in NumPy — plus `mlscratch`, a pip-installable package (`pip install scratchkit`) with a consistent, scikit-learn-style API and 1,100+ tests.**
+
+[![CI](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/actions/workflows/ci.yml/badge.svg)](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/actions)
+[![PyPI](https://img.shields.io/pypi/v/scratchkit.svg)](https://pypi.org/project/scratchkit/)
+[![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/LICENSE)
+[![Python 3.10+](https://img.shields.io/badge/Python-3.10+-blue.svg)](https://python.org)
+[![Stars](https://img.shields.io/github/stars/Mattral/ML-AI-Algorithms-from-scratch?style=social)](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/stargazers)
+
+> **What's here:** readable, standalone implementations of algorithms you already know by name, written to show the math in code, not to be fast.
+>
+> **What's new:** `src/mlscratch/` — a pip-installable package with `fit()`/`predict()`/`transform()` APIs, full type hints, and a test suite that cross-checks correctness against scikit-learn wherever a reference implementation exists.
+
+---
+
+## What makes this different from the dozens of similar repos
+
+There are many "ML from scratch" repos on GitHub. The honest differentiators here:
+
+- **Bayesian methods are first-class.** Most from-scratch repos stop at supervised learning + neural nets. This one includes Bayesian Neural Networks, Gaussian Processes, Hidden Markov Models, Bayesian Networks, and Kalman Filters — algorithms most tutorials skip because they're harder to implement correctly.
+- **RL goes beyond DQN.** DDPG, TD3, SAC, and PPO are included alongside tabular Q-Learning and DQN — non-trivial to implement correctly from scratch, and rare to see done well in a single repo.
+- **The `src/mlscratch` package is real, not a wrapper.** Every estimator is implemented in pure NumPy — no calling out to scikit-learn at runtime. scikit-learn only appears in the *test suite*, as a correctness oracle, never as a dependency of the library itself.
+- **Kernel SVM via real SMO, gradient boosting with proper Newton-step leaves, multiclass-native AdaBoost (SAMME.R)** — the ensemble/kernel methods aren't toy simplifications; several are verified to match scikit-learn's output to floating-point tolerance on real benchmarks.
+
+---
+
+## Quick start
+
+### Browse the standalone scripts (no install needed)
+
+```bash
+git clone https://github.com/Mattral/ML-AI-Algorithms-from-scratch
+cd ML-AI-Algorithms-from-scratch
+
+pip install numpy matplotlib scikit-learn   # only deps, for the standalone scripts
+
+python "Supervised/LinearRegression/linear_regression.py"
+python "Neural Networks/Transformer/transformer.py"
+python "Reinforcement/PPO/ppo.py"
+```
+
+### Use the package
+
+```bash
+pip install scratchkit                # from PyPI — the import name is still `mlscratch`
+```
+
+```bash
+# — or, for local development —
+pip install -e .                  # installs src/mlscratch in editable mode
+# pip install -e ".[dev]"         # + pytest, ruff, black, mypy, for development
+
+pytest tests/ -v                  # run the test suite
+python -m mlscratch info          # package + sub-package summary
+python -m mlscratch list supervised
+```
+
+```python
+from mlscratch.supervised import RandomForestClassifier
+from mlscratch.preprocessing import StandardScaler, train_test_split
+from mlscratch.metrics import classification_report
+
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, stratify=y)
+
+scaler = StandardScaler().fit(X_train)
+model = RandomForestClassifier(n_estimators=200, max_depth=6, oob_score=True)
+model.fit(scaler.transform(X_train), y_train)
+
+print(f"OOB score: {model.oob_score_:.3f}")
+print(classification_report(y_test, model.predict(scaler.transform(X_test))))
+```
+
+See [`examples/`](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/tree/main/examples) for six runnable end-to-end scripts covering decision trees, random forests, kernel SVMs, gradient boosting, AdaBoost, and a full no-sklearn classification + regression pipeline.
+
+---
+
+## What's implemented
+
+### `mlscratch` package (`src/mlscratch/`)
+
+| Sub-package | Contents | Tests |
+|---|---|---|
+| `mlscratch.supervised` | Linear/Ridge/Lasso/ElasticNet/Logistic regression, KNN, **DecisionTree** (classifier + regressor), **RandomForest** (bagging + OOB scoring), kernel **SVC** (SMO; linear/poly/rbf/sigmoid, one-vs-rest multiclass), **GradientBoosting** (classifier + regressor, squared/absolute-error loss), **AdaBoost** (SAMME / SAMME.R, multiclass-native) | 162 |
+| `mlscratch.unsupervised` | K-Means++, K-Medoids, DBSCAN, Agglomerative Clustering, PCA, t-SNE, FastICA, Gaussian Mixture Model (EM), Apriori | 120 |
+| `mlscratch.bayesian` | Naive Bayes (Gaussian/Multinomial/Bernoulli), Bayesian Linear Regression, Bayesian Network, Bayesian Neural Network (mean-field VI), Gaussian Process Regression, Hidden Markov Model, Kalman Filter | 171 |
+| `mlscratch.reinforcement` | Q-Learning, Double Q-Learning, DQN (Double + Dueling + PER), DDPG, TD3, PPO (GAE-λ), SAC, plus shared `GridWorld`/`ReplayBuffer`/`PrioritizedReplayBuffer` utilities | 218 |
+| `mlscratch.neural` | Single/Multi-Layer Perceptron, Autoencoder (vanilla/denoising/variational), RNN/LSTM/Encoder-Decoder, a small CNN (Conv2D/Pool/BatchNorm), Attention + Transformer encoder, GAN, Hopfield Network, Restricted Boltzmann Machine, RBF Network, Complex-Valued NN | 372 |
+| `mlscratch.metrics` | accuracy/precision/recall/F1, confusion matrix, `classification_report`, ROC/AUC, log loss, MSE/RMSE/MAE/MAPE, R², explained variance — every metric checked against scikit-learn | 48 |
+| `mlscratch.preprocessing` | StandardScaler, MinMaxScaler, RobustScaler, Normalizer, LabelEncoder, OneHotEncoder, PolynomialFeatures, `train_test_split` (with stratification) | 62 |
+
+**1,153 tests total.** A handful (~18) fail under the newest NumPy/SciPy releases in this environment due to upstream API drift in unrelated modules (Bayesian networks, reinforcement learning buffers, ICA) — tracked as known issues, not part of this release's scope.
+
+### Standalone scripts (original, by category)
+
+These are the original from-scratch scripts the package above was distilled from — browse them like a reference, run them directly, no install required.
+
+- **`Supervised/`** — Linear/Ridge/Lasso Regression, Logistic Regression, k-NN, Decision Trees, Random Forest, Naive Bayes, SVM
+- **`Unsupervised/`** — K-Means++, K-Medoids, DBSCAN, Hierarchical Clustering, PCA, t-SNE, ICA, Gaussian Mixture Model, EM, Self-Organising Map, Apriori
+- **`Neural Networks/`** — Single/Multi-Layer Perceptron, Simple RNN, LSTM, Simple CNN, Encoder-Decoder, Self-Attention, Transformer, Autoencoder, GAN, Boltzmann Machine, Hopfield Network, RBF Networks
+- **`Reinforcement/`** — Q-Learning, DQN, DDPG, PPO, SAC
+- **`Bayesian Learning/`** — Bayesian Inference, Bayesian Linear Regression, Bayesian Network, Bayesian Neural Networks, Gibbs Sampling, Metropolis-Hastings, Variational Inference
+
+---
+
+## Design philosophy
+
+Every implementation applies the same principles:
+
+- Explicit loops over vectorised one-liners when clarity improves
+- Model logic, loss computation, and parameter updates in separate functions
+- The package layer (`src/mlscratch`) calls **only** NumPy at runtime — scikit-learn appears solely in the test suite, as a correctness oracle
+- Short files: most standalone scripts are 100–300 lines; package modules favor one well-documented class per concern
+
+**This trades raw performance for readability and correctness-by-inspection. That's intentional.**
+
+If you're looking for production-speed implementations, use scikit-learn, PyTorch, or JAX. If you want to read the math in code form — or verify it against a reference implementation in the test suite — this is the repo.
+
+---
+
+## Recommended learning path
+
+If you're working through this systematically:
+
+1. Start with `Supervised/LinearRegression` (or `mlscratch.supervised.LinearRegression`) — the simplest possible end-to-end example
+2. Move to `LogisticRegression` — same structure, adds sigmoid + cross-entropy
+3. Then `DecisionTreeClassifier` → `RandomForestClassifier` → `GradientBoostingClassifier`/`AdaBoostClassifier` — the tree-ensemble family, building on a shared CART implementation
+4. Then `Neural Networks/SingleLayerPerceptron` → `MultiLayerPerceptron` — backprop from first principles
+5. Then any of: Unsupervised (PCA → GMM → t-SNE), Reinforcement (Q-Learning → DQN → PPO/SAC), or Bayesian (Naive Bayes → Bayesian Linear Regression → Variational Inference)
+
+Each folder/module is reasonably self-contained — jump to any algorithm without reading the others first.
+
+---
+
+## Repository layout
+
+```
+ML-AI-Algorithms-from-scratch/
+│
+├── Supervised/              Standalone scripts: LinearRegression, SVM, etc.
+├── Unsupervised/            Standalone scripts: KMeans++, DBSCAN, t-SNE, etc.
+├── Neural Networks/         Standalone scripts: MLP, LSTM, Transformer, GAN, etc.
+├── Reinforcement/           Standalone scripts: DQN, DDPG, PPO, SAC, etc.
+├── Bayesian Learning/       Standalone scripts: BNN, VI, MCMC, etc.
+│
+├── src/mlscratch/           Pip-installable package
+│   ├── supervised/          Linear models, KNN, trees, ensembles, kernel SVM
+│   ├── unsupervised/        Clustering, dimensionality reduction, association rules
+│   ├── bayesian/            Naive Bayes, BLR, BNN, GP, HMM, Bayesian Networks, Kalman
+│   ├── reinforcement/       Q-Learning, DQN, DDPG, TD3, PPO, SAC
+│   ├── neural/              Perceptrons, autoencoders, RNN/CNN, attention, GAN, ...
+│   ├── metrics/             Classification & regression evaluation metrics
+│   └── preprocessing/       Scalers, encoders, polynomial features, train_test_split
+│
+├── examples/                Runnable end-to-end scripts (no sklearn at runtime)
+├── tests/                   1,153 tests, mirroring the src/mlscratch layout
+├── docs/                    Roadmap (MkDocs site planned, see roadmap.md)
+├── pyproject.toml           Package metadata + deps
+├── CHANGELOG.md             Keep-a-Changelog formatted release history
+├── roadmap.md               P0 / P1 / P2 backlog
+├── .github/workflows/       CI: lint → test matrix → build → PyPI release
+└── README.md
+```
+
+---
+
+## Contributing
+
+The most useful contributions right now:
+
+- **Add a standalone script** for an algorithm not yet covered (check the folder first)
+- **Port a standalone script** into `src/mlscratch` with a matching test file in `tests/`
+- **Fix a numerical issue** — some implementations have known edge cases under newer NumPy/SciPy releases (see the known-issues note above; open an issue or PR)
+
+Standard flow: fork → branch → PR. CI runs `ruff`, `black --check`, and the full `pytest` suite on every PR. See [`CONTRIBUTING.md`](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/CONTRIBUTING.md) for the full guide, and [`roadmap.md`](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/roadmap.md) for what's planned next.
+
+---
+
+## Honest scope
+
+The standalone scripts under `Supervised/`, `Neural Networks/`, etc. are a **learning reference**, not a performance library: some use toy datasets, a few have hardcoded hyperparameters to keep the code short, and none are tuned for speed at scale.
+
+The `src/mlscratch` package is more rigorous (typed, tested, cross-checked against scikit-learn) but is still pure-Python/NumPy — it will not outrun scikit-learn or XGBoost on large datasets, and that was never the goal. The public API is stabilising but may still change between minor versions before a 1.0 release; pin a version if you're building on top of it.
+
+---
+
+## License
+
+Apache 2.0 — see [LICENSE](https://github.com/Mattral/ML-AI-Algorithms-from-scratch/blob/main/LICENSE).