mxagar/machine_learning_coursera

My notes on the Coursera course by Andrew Ng "Machine Learning".

Machine Learning

These are my notes on the Coursera course by Andrew Ng "Machine Learning".

The course is divided in 11 learning weeks which cover several topics of Machine Learning. I divided the course in topic folders, which do not necessarily match the learning weeks. Each folder contains my notes on the topic as well as the course material: slides, scripts, etc.:

• 01_LinearRegression/
• 02_LogisticRegression/
• 03_NeuralNetworks/
• 04_MLSystemDesign/
• 05_SupportVectorMachines/
• 06_UnsupervisedLearning/
• 07_Anomaly_Recommender/: Anomaly Detection and Recommender Systems
• 08_OCR_ApplicationExample/

Additionally, the exercises are located in the folder exercises/ex1-ex8-octave.

I followed the original course by Andrew Ng, which has exercises in Octave/Matlab, but I completed the exercises in Jupyter notebooks running on a conda environment. Alternatively, if you'd like to implement the exercises in Python with Numpy, you can check this repository: ml-coursera-python-assignments.

Setup

In order to run my implementations from the Jupyter notebooks, you need to install the octave kernel in a conda environment. The following is the list of steps I followed to set things up on my Mac:

# 1) Install brew: https://brew.sh/
# 2) Install octave with brew
brew install octave
# 3) Create and activate a conda environment
conda create -n ml-octave python=3.6
conda activate ml-octave
# 4) Add the necessary packages to your environment
conda config --add channels conda-forge
conda install octave_kernel
conda install texinfo # For the inline documentation (shift-tab) to appear.

Authorship

Mikel Sagardia, 2022.
No guarantees.

If you find this repository helpful and use it, please refer back to the original source.