ml-from-scratch-2019

Labs

• 0. Python basics
• 1. NumPy basics
• 2. Logistic Regression (Starter)
• 3. Logistic Regression Part II (starter)
• 4. Linear Regression (Starter)
• 5. Decision Trees (Starter)
• 5. Decision Trees (Complete)

Slides

• Intro to ML (by examples)
• Ethics and Data Science (Fast.ai)
• ML Applications (Fast.ai)

Recommended courses

• Machine Learning (Coursera)
• Introduction to Machine Learning for Coders (Fast.ai)
• Practical Deep Learning For Coders

Resources

• The State of Data Science & Machine Learning (2017)
• Fast.ai - International Fellowship
• Mathematics for Machine Learning (Book)

Exam

Lowest error solution

• By team “А. К., М. М., Д.И.”

Challenge

• Kaggle Challenge

Requirements for entering the exam

• Implementation (from scratch) of a Random Forest model
• Train your model on the training data, predict on the test data and upload the result to Kaggle

Decision tree source code

Our implementation of a Decision tree model can be found here - Decision Tree source code. You can use it to build your Random Forest.

The notebook contains example code for generating your submission.csv file that you must upload for Kaggle to score it.

Resources

• Introduction to Machine Learning for Coders
• @williamkoehrsen/random-forest-simple-explanation-377895a60d2d">Random Forest Simple Explanation
• Random Forest in Python

Teams

• B.H., L.M.
• A.R., L.V.
• G.M., G.F., S.O.
• L.P. M.P.
• Кьополу - L.B., E.K.
• Д.А., Т.М.
• Мусака - Т.Н., З.К.
• А. К., М. М., Д.И.
• Д.М., А.П., В.К.
• Е.Д., Д.Д.
• Г.П., Й. И.