mlcourse.ai- September 2, 2019.
mlcourse.ai is an open Machine Learning course by OpenDataScience.The course is designed to perfectly balance theory and practice. You can take part in several Kaggle Inclass competitions held during the course.
Author: Arina Lopukhova (@erynn). Edited by Yury Kashnitskiy (@yorko) and Vadim Shestopalov (@vchulski). You can ask your questions in two special threads ⬇️ Vadim @vchulski. you can freely write to #mlcourse_ai BUT please use threads :thread-please: https://github.com/Yorko/mlcourse.ai/issues https://github.com/Yorko/mlcourse.ai
To follow the course schedule and all deadlines, calender : https://bit.ly/2HB9378
you watch lectures on your own, solve quizzes (mostly based on the material in articles), then we'll have several live sessions (YouTube as well) where we discuss questions in quizzes.
1st quiz will be published on September 9, follow announcements in the #mlcourse_ai_news channel.
Syllabus
- Module 1. Exploratory Data Analysis, 1 assignment
- Module 2. Decision trees, Random Forest & gradient boosting. 1 assignment in a form of a competition, 1 quiz
- Module 3. Linear classification & regression models. 1 assignment, 1 competition, 1 quiz
- Module 4. Unsupervised learning, time series, Vowpal Wabbit. 2 assignments, one of them in a form of a competition
-
- 1 more competition held throughout the course
Navigating this site: https://mlcourse.ai
- on this page, you’ll find the main course content, see icons below
- lectures(https://mlcourse.ai/lectures)- video recordings of fall 2018 lectures
- prerequisites(https://mlcourse.ai/prerequisites) - what you need to join this course
- assignments(https://mlcourse.ai/assignments) - links to demo assignments
- tutorials(https://mlcourse.ai/tutorials) - tutorials written by course participants
- resources(https://mlcourse.ai/resources) - a list of all materials constituting the course
- rating(https://mlcourse.ai/rating) - current rating and top-performers of previous sessions
- FAQ(https://mlcourse.ai/faq) - Frequently Asked Questions
- contacts(https://mlcourse.ai/contacts) - how to reach the course team
- contrib (https://mlcourse.ai/contrib)- how you can help
This YouTube playlist contains fall 2018 video lectures. A couple more recordings will be added in fall 2019 session.
Introduction - video https://youtu.be/QKTuw4PNOsU, slideshttps://bit.ly/2NuadRV
- Exploratory data analysis with Pandas - https://youtu.be/fwWCw_cE5aI
- Visualization, main plots for EDA - https://www.youtube.com/watch?v=WNoQTNOME5g
- Decision trees - theory(https://youtu.be/H4XlBTPv5rQ) and practical(https://youtu.be/RrVYO6Td9Js) part
- Logistic regression - theoretical foundations(https://www.youtube.com/watch?v=l3jiw-N544s), practical(https://www.youtube.com/watch?v=7o0SWgY89i8) part (baselines in the “Alice” competition)
- Ensembles and Random Forest - part 1(https://www.youtube.com/watch?v=neXJL-AqI_c). Classification metrics - part 2(https://www.youtube.com/watch?v=aBOMYqGUlWQ). Example of a business task, predicting a customer payment - part 3
- Linear regression and regularization - theory(https://youtu.be/ne-MfRfYs_c), LASSO & Ridge, LTV prediction - practice(https://youtu.be/B8yIaIEMyIc)
- Unsupervised learning - Principal Component Analysis(https://youtu.be/-AswHf7h0I4) and Clustering(https://youtu.be/eVplCo-w4XE)
- Stochastic Gradient Descent for classification and regression - part 1(https://youtu.be/EUSXbdzaQE8), part 2 TBA
- Time series analysis with Python (ARIMA, Prophet) - video(https://youtu.be/_9lBwXnbOd8)
- Gradient boosting: basic ideas - part 1(https://youtu.be/g0ZOtzZqdqk), key ideas behind Xgboost, LightGBM, and CatBoost + practice - part 2(https://youtu.be/V5158Oug4W8)
Outroduction:( https://youtu.be/FrIW8ixKakw)(https://bit.ly/2s0sjD7)