Machine Learning Concepts

Author : Ramin Anushiravani

Update:

      Sep 10th , 2023
         
      July 4th , 2024

Implementation of common ML algorithms in Numpy or Pytorch. This is purley for educational purposes. All notebooks should run propely. I tried my best to implement most of the following models from scratch using numpy or pytroch.

Models and Algorithms

Regression

* Linear regression   

       Linear_Regression.ipynb , Linear and Non-Linear Regression Pytorch.ipynb

* Logistic regression 

        Logistic_Regression.ipynb , Logistic_Regression_SimpleNN_Pytorch.ipynb , Logistic_Regression_Pytorch.ipynb

Trees

* Decision tree and Random forest algorithm

    decision_tree-random-forest-pytorch.ipynb
    
* Gradient Boosting algorithm

        gradient_boosting-pytorch.ipynb 

SVM

* SVM algorithm

        svm-pytorch.ipynb

Naive Bayes

* Naive Bayes algorithm

        Naive Bayes.ipynb

KNN

* KNN algorithm  

        k-nearest-pytorch.ipynb

K means

* K-means  

        k-means_pytorch.ipynb , k-means_numpy.ipynb

Dim Reduce

* Dimensionality Reduction Algorithms

        pca.ipynb

Neural

* Backpropagation

        backprop_pytorch.ipynb

* RNN  

        TBD

* CNN 

        TBD

Tasks

* Text Classification 

        Pytorch-Classification.ipynb

* Building a Recommendation System

        Recommendation.ipynb

Disclaimer

I used ChatGPT to create definition of the model. You can see the snapshots added to the notebooks. In some cases I used ChatGPT to generate part of the code as well.