https://github.com/dzhokn/machine-learning-from-scratch-without-libraries

Backprop, NNs, Regression models, Classification models, etc. - all writtern from scratch without PyTorch or TensorFlow
https://github.com/dzhokn/machine-learning-from-scratch-without-libraries

from-scratch machine-learning no-libraries

Last synced: 7 months ago
JSON representation

Backprop, NNs, Regression models, Classification models, etc. - all writtern from scratch without PyTorch or TensorFlow

• Host: GitHub
• URL: https://github.com/dzhokn/machine-learning-from-scratch-without-libraries
• Owner: dzhokn
• License: mit
• Created: 2025-06-10T07:15:55.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-10-03T11:13:48.000Z (10 months ago)
• Last Synced: 2025-10-03T13:16:19.675Z (10 months ago)
• Topics: from-scratch, machine-learning, no-libraries
• Language: Jupyter Notebook
• Homepage:
• Size: 27.1 MB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Machine learning from scratch without libraries

Backpropagation, Neural Networks, Regression models, Classification models, CNNs, SVM, Trees, Forests, etc. - all writtern from scratch without PyTorch or TensorFlow.

Altough, I am briefly covering the theory (and mathematics) behind each of the models, it is required for the reader to already have some basic knowledge in machine learning and the models in question.

![Machine Learning Roadmap](ml_roadmap.png)

NB: I will be updating this roadmap on the fly - after adding new and new models.