showml 1.6.12

A Python package of Machine Learning Algorithms implemented from scratch

---

A Python package of Machine Learning Algorithms implemented from scratch.

The aim of this package is to present the working behind fundamental Machine Learning algorithms in a transparent and modular way.

NOTE: The implementations of these algorithms are not thoroughly optimized for high computational efficiency.

📝 Table of Contents

• Getting Started
• Contents
• Contributing
• License

🏁 Getting Started

To install the package directly from PyPi:

$ pip install showml

To clone the repository and view the source files:

$ git clone https://github.com/hasnainroopawalla/ShowML.git
$ cd ShowML
$ pip install -r requirements.txt

Remember to add ShowML/ to the PYTHONPATH environment variable before using locally:-

• For Windows:

  $ set PYTHONPATH=%PYTHONPATH%;<path-to-directory>\ShowML
  

• For MacOS:

  $ export PYTHONPATH=/<path-to-directory>/ShowML:$PYTHONPATH
  

• For Linux:

  $ export PYTHONPATH="${PYTHONPATH}:/<path-to-directory>/ShowML"
  

Check out: showml/examples/

📦 Contents

ShowML currently includes the following content, however, this repository will continue to expand in order to include implementations of many more Machine Learning Algorithms.

Models

• Linear

  • Linear Regression (showml.linear_model.regression.LinearRegression)
  • Logistic Regression (showml.linear_model.regression.LogisticRegression)

• Non-Linear

  • Sequential (showml.deep_learning.model.Sequential)

Deep Learning

• Layers

  • Dense (showml.deep_learning.layers.Dense)

• Activations

  • Sigmoid (showml.deep_learning.activations.Sigmoid)
  • ReLu (showml.deep_learning.activations.Relu)
  • Softmax (showml.deep_learning.activations.Softmax)

Optimizers

• Stochastic/Batch/Mini-Batch Gradient Descent (showml.optimizers.SGD)
• Adaptive Gradient (showml.optimizers.AdaGrad)
• Root Mean Squared Propagation (showml.optimizers.RMSProp)

Loss Functions

• Mean Squared Error (showml.losses.MeanSquaredError)
• Binary Cross Entropy (showml.losses.BinaryCrossEntropy)
• Categorical Cross Entropy (showml.losses.CrossEntropy)

✏️ Contributing

1. Fork the repository.
2. Commit and push your changes to your own branch.
3. Install and run the necessary housekeeping dependencies (pre-commit, mypy and pytest):

   $ pip install pre-commit mypy pytest
   

4. Run these housekeeping checks locally and make sure all of them succeed (required for the CI to pass):-

   $ pre-commit run -a
   $ mypy .
   $ pytest
   

5. Open a Pull Request and I'll review it.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.