ML

This repo contains personal/pet projects made for understanding algorithms from scratch and training models in their entirety without using machine learning libraries for sequential layers or feature classification

Projects

1. Spam Email Classfier Using SVM
2. Movie Recommendation System Using Collaborative_Filtering
3. Image Compression using K-means Algorithm
4. Facial images dimensionality reduction and reconstruction using PCA
5. Anomaly Detection Algorithm
6. Handwriting recognition using Neural Networks Backprogapation
7. Handwriting recognition using Neural Network in MATLAB Pattern Recognition App
8. Handwriting recognition using One-vs-All Classification Neural Networks
9. Handwriting recognition using One-vs-All Logisitc Regression
10. Understanding models with different bias-variance properties using learning curves and k-folds cross validation
11. Regularized Logistic Regression
12. Unregularized Logistic Regression
13. Multivariate Linear Regression for Housing Prediction
14. Univariate Linear Regression for Cost Prediction
15. MATLAB Onramp Projects :
    • Stellar Motion Determination
    • Data Visualization
    • Graph visualization and Audio Signal Processing

How to run?

You can run project either in octave or MATLAB.

1. Clone repository using git clone
2. cd to project directory and either run following command in octave or MATLAB
3. run('script_name.m') to run the project

Where to find help?

• If you do not have Octave installed, please refer to the installation instructions on the Octave Download official site.
• At the Octave/MATLAB command line, typing help followed by a function name displays documentation for a built-in function. For example, help plot will bring up help information for plotting. Further documentation for Octave functions can be found at the Octave documentation pages.
• MATLAB is a proprietary software but see if your school/university has a MATLAB campus license.
• MATLAB documentation can be found at the MATLAB documentation pages.