Glagan/42-ft_linear_regression

The ft_linear_regression project at 42

ft_linear_regression

Requirements

Python 3 and matplotlib.

Usage

Train to get theta0 and theta1 first:

python train.py

You can then estimate a price for a given mileage:

python estimate.py [mileage]

To see the effect of changing the learning rate with the given parameters you can use learning_rate.py.
This will display a graph with the sum of squarred residuals for 1000 iterations.

python learning_rate.py

Bonuses

• Graph with dataset and trained linear regression line
• Graph with the effect of different learning rates

Resources

• Machine Learning
  • Full Course
    • This is mandatory
    • Look at the 42-AI Bootcamp Machine Learning module 6 to see what parts you should watch
    • https://www.coursera.org/learn/machine-learning
  • StatQuest
    • https://www.youtube.com/playlist?list=PLblh5JKOoLUICTaGLRoHQDuF_7q2GfuJF
    • Linear Regression
      • https://www.youtube.com/watch?v=PaFPbb66DxQ
    • Gradient Descent
      • https://www.youtube.com/watch?v=sDv4f4s2SB8
  • Linear Regression
    • https://www.youtube.com/watch?v=GhrxgbQnEEU
    • https://towardsdatascience.com/linear-regression-cost-function-gradient-descent-normal-equations-1d2a6c878e2c
  • Linear Regression with Gradient Descent
    • https://towardsdatascience.com/linear-regression-using-gradient-descent-97a6c8700931
  • Feature Scaling
    • https://en.wikipedia.org/wiki/Feature_scaling#Standardization
    • https://stackoverflow.com/questions/32108179/linear-regression-normalization-vs-standardization
    • https://stackoverflow.com/a/59016254/7794671 with https://ibb.co/178qWcQ
• 42-AI
  • https://github.com/42-AI/bootcamp_machine-learning
• Python
  • https://github.com/42-AI/bootcamp_python
  • Head First Python, Second Edition (Paul Barry, Editions O'Reilly)