explaingit

dibgerge/ml-coursera-python-assignments

5,566Jupyter NotebookAudience · researcherComplexity · 1/5Setup · easy

TLDR

Python rewrites of every Andrew Ng Coursera machine learning assignment as interactive Jupyter Notebooks, fully compatible with the original Coursera grader for automatic scoring.

Mindmap

mindmap
  root((repo))
    What it does
      Python ML assignments
      Replaces MATLAB versions
      Coursera grader support
    Topics covered
      Linear regression
      Logistic regression
      Neural networks
      SVM and clustering
    Setup options
      Anaconda local install
      Deepnote online
    Tech Stack
      Python
      Jupyter Notebooks
      NumPy and SciPy
      Matplotlib
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Complete Andrew Ng Coursera ML assignments in Python instead of MATLAB or Octave.

USE CASE 2

Submit completed notebooks to the original Coursera grader for automatic scoring and a certificate.

USE CASE 3

Practice machine learning algorithms like linear regression, neural networks, and SVM using Python.

USE CASE 4

Run all assignments in Deepnote online without installing any software locally.

Tech stack

PythonJupyterNumPySciPyMatplotlibAnaconda

Getting it running

Difficulty · easy Time to first run · 30min

Requires Python 3 with NumPy, SciPy, and Matplotlib, use Anaconda or Deepnote for the simplest setup.

License terms were not stated in the README.

In plain English

This repository contains Python versions of the programming assignments from Andrew Ng's Machine Learning course on Coursera, one of the most widely taken introductory courses on the subject. The original course was designed around MATLAB and Octave, two older scientific computing environments. Since Python has become the standard language for machine learning work in recent years, this project re-wrote every assignment in Python so students can practice with tools they are more likely to use in real projects. Each assignment lives in its own folder and is written as a Jupyter Notebook. A Jupyter Notebook is an interactive document where you can read instructions, write code, and run it all in the same place inside a web browser. The instructions in these notebooks have been rewritten from scratch to reference Python instead of MATLAB, so there is no need to refer back to the original course materials. A notable feature is that the assignments can be submitted to the original Coursera grader for automatic scoring, the same grader used for the MATLAB versions. After finishing each section of an assignment, a cell in the notebook prompts the student to submit their work for a grade. The original course structure and grading flow are preserved. Setting up the environment requires installing Python 3 along with several scientific libraries: NumPy for numerical computing, SciPy for math utilities, and Matplotlib for plotting. The README recommends installing everything through Anaconda, a package manager that bundles Python with these scientific tools and simplifies setup on Windows and Linux alike. A provided environment file lets students create the right configuration in one command. For students who prefer not to install anything locally, the assignments can also be run in Deepnote, an online environment that runs notebooks in a browser with no local setup required. Links to Python tutorials for beginners are included in the README for those who are new to the language or to Jupyter Notebooks.

Copy-paste prompts

Prompt 1
I am starting Andrew Ng Coursera ML in Python. How do I set up the Anaconda environment to run these notebooks?
Prompt 2
How do I submit my completed Jupyter Notebook assignment to the Coursera grader for automatic scoring?
Prompt 3
Show me how the Python code in the logistic regression assignment compares to the original MATLAB version.
Prompt 4
Which exercises in ml-coursera-python-assignments cover neural networks, and what Python libraries do they use?
Prompt 5
I get an import error running one of the notebooks in Deepnote. How do I fix a missing library?
Open on GitHub → Explain another repo

← dibgerge on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.