code-academy-ml

Machine Learning classes from code academy. (https://www.codecademy.com/programs/b8e789be014182f17868650a1556c946)

Grade: 100% (see julien-amar#1)

Regression

LinearRegression: a model for understanding the relationship between an input and output numerical variables using a linear regression: a * x + b

Multiple Linear Regression: Similar to LinearRegression, but enables to understand the relationship between multiple inputs and output numerical variables using a linear regression: a1 * x1 + a2 * x2 + ... + an * xn + b

Classifier

Naive Bayes: Supervised machine learning algorithm that leverages Bayes' Theorem (probabilistic approach) to make predictions and classifications.

K-Nearest Neighbors (KNN): Non-parametric method used for classification. The input consists of the k closest training examples in the feature space.

K-Nearest Neighbor Regressor (KNNR): Similar to KNN, but output the average of the values of its k nearest neighbors for a specific feature.

Support Vector Machines (SVM): Supervised machine learning model, defining a decision boundary and then seeing what side of the boundary an unclassified point falls on. SVMs can efficiently perform a non-linear classification using what is called the kernel trick, implicitly mapping their inputs into high-dimensional feature spaces.

Clustering

K-Mean: Unsupervised machine learning model, using a defined number of cluster (k) and randomly placed centroid to cluster data. Inertia define the distance with associated centroids (lower is better).

K-Means++: K-Means++ is an updated version of K-Means that aims to avoid the convergence and clustering problems by placing the initial centroids in a more effective way.

Plan

• Introduction, Setup, & Prework

  • Welcome!
  • Learning Experience
  • Setting Up
  • Slack
  • Project Review
  • Certification
  • Optional Prework

• Unit 1: What is Machine Learning?

  • WEEK 1, DAY 1
  • Lesson: Why Machine Learning?
  • Article: Supervised vs. Unsupervised
  • WEEK 1, DAY 2
  • Article: The Machine Learning Process
  • Article: Scikit-Learn Cheatsheet

• Unit 2: Regression

  • WEEK 1, DAY 3
  • Lesson: Distance Formula
  • WEEK 1, DAYS 4 & 5
  • Article: Regression vs. Classification
  • Lesson: Linear Regression Lesson
  • Quiz: Linear Regression Quiz
  • WEEK 1, DAY 6
  • Article: Training Set vs Validation Set vs Test Set
  • Lesson: Multiple Linear Regression (StreetEasy)
  • Quiz: Multiple Linear Regression Quiz
  • WEEK 2, DAY 1
  • Project: Linear Regression for the Bees
  • WEEK 2, DAY 2
  • Lesson: Accuracy, Precision and Recall
  • Quiz: Accuracy, Precision and Recall Quiz
  • WEEK 2, DAYS 3 & 4
  • Yelp Rating Predictor Cumulative Project
  • Article: Yelp Dataset Terms of Use
  • WEEK 2, DAYS 5 & 6
  • Catch-Up and Reflect

• Unit 3: Classification

  • WEEK 3, DAY 1
  • Article: Normalization
  • WEEK 3, DAY 2
  • Lesson: Naive Bayes Classifier
  • Quiz: Naive Bayes Classifier Quiz
  • WEEK 3, DAY 3
  • Project: Naive Bayes Classifier Project
  • WEEK 3, DAY 4
  • Lesson: K-Nearest Neighbors
  • Quiz: K-Nearest Neighbors Quiz
  • WEEK 3, DAY 5
  • Lesson: K-Nearest Neighbors Regression
  • WEEK 3, DAY 6
  • Project: Breast Cancer Classification
  • Article: The Dangers of Overfitting
  • WEEK 4, DAY 1
  • Lesson: Support Vector Machine
  • Quiz: Support Vector Machine Quiz
  • WEEK 4, DAY 2
  • Project: Sports Vector Machine
  • WEEK 4, DAYS 3 & 4
  • Twitter Classification Cumulative Project
  • WEEK 4, DAYS 5 & 6
  • Catch-Up and Reflect

• Unit 4: Unsupervised Learning

  • WEEK 5, DAY 1
  • Lesson: K-Means Clustering
  • Quiz: K-Means Clustering Quiz
  • WEEK 5, DAY 2
  • Lesson: K-Means++ Lesson
  • WEEK 5, DAY 3
  • Project: Handwriting Recognition Project
  • WEEK 5, DAYS 4 & 5
  • Masculinity Survey Cumulative Project
  • WEEK 5, DAY 6
  • Catch-Up and Reflect

• Unit 5: Neural Network Teaser

  • WEEK 6, DAY 1
  • Article: What are Neural Networks?
  • WEEK 6, DAY 2
  • Lesson: Perceptron Lesson
  • Quiz: Perceptron Quiz
  • WEEK 6, DAY 3
  • Project: Perceptron Project

• Unit 6: Capstone Project

  • REST OF WEEK 6 AND WEEK 7
  • Project: Date-A-Scientist

Useful links

• Numpy and Scipy Documentation: Library for processing data (as arrays)
• Pandas: Python Data Analysis Library
• Metaplotlib: Python plotting library
• scikit learn: Machine Learning library in Python