Kaggle Mini Courses - Time Series Part 1/2
Event page: https://www.meetup.com/Deep-Learning-...
Join us for a new adventure on our journey to deep learning and data science in general 🎉 powered by Kaggle!
This brand new course is free, and you can now earn a certificate. Just bring your curiosity and get ready to meet our growing community 😀
Agenda:
Introductions and get to know our community
Deep Learning Adventures - Coding Presentation:
We will be following the course material and exercises at:
https://www.kaggle.com/learn/time-series
1 Linear Regression With Time Series
Use two features unique to time series: lags and time steps.
2 Trend
Model long-term changes with moving averages and the time dummy.
3 Seasonality
Create indicators and Fourier features to capture periodic change.
4 Time Series as Features
Predict the future from the past with a lag embedding.
5 Hybrid Models
Combine the strengths of two forecasters with this powerful technique.
6 Forecasting With Machine Learning
Apply ML to any forecasting task with these four strategies.
Bonus content:
Bonus 1: Facebook/Meta Prophet
Prophet is a forecasting procedure implemented in R and Python. It is fast and provides completely automated forecasts that can be tuned by hand by data scientists and analysts.
URL: https://facebook.github.io/prophet
Bonus 2: ROCKET
ROCKET [1] transforms time series using random convolutional kernels (random length, weights, bias, dilation, and padding). ROCKET computes two features from the resulting feature maps: the max, and the proportion of positive values (or ppv). The transformed features are used to train a linear classifier.
[1] Dempster A, Petitjean F, Webb GI (2019) ROCKET: Exceptionally fast and accurate time series classification using random convolutional kernels. arXiv:1910.13051
URL: https://www.sktime.org/en/stable/exam...
MINIROCKET: A Very Fast (Almost) Deterministic Transform for Time Series Classification arXiv:2012.08791
https://github.com/angus924/minirocket
Bonus 3: The M5 Forecasting Competition
The aim of the M5 Competition is similar to the previous four: that is to identify the most appropriate method(s) for different types of situations requiring predictions and making uncertainty estimates. Its ultimate purpose is to advance the theory of forecasting and improve its utilization by businesses and non-profit organizations. Its other goal is to compare the accuracy/uncertainty of ML and DL methods vis-à-vis those of standard statistical ones, and assess possible improvements versus the extra complexity and higher costs of using the various methods.
URLs:
https://mofc.unic.ac.cy/m5-competition
https://www.kaggle.com/c/m5-forecasti...
https://eng.uber.com/m4-forecasting-c...
https://www.innovating-automation.blo...
https://www.researchgate.net/publicat...
https://github.com/Mcompetitions/M5-m...
Deep Learning YouTube playlists, feel free to share and subscribe 😀
https://bit.ly/dla-youtube
The recording of this cool event 😎 is available at:
http://bit.ly/dla-kaggle-courses
Join us on Slack:
https://join.slack.com/t/deeplearning...
Spread the word about our meetup 🎉
Are you excited to join us? See you soon!
Best
Robert, David and George
Watch video Kaggle Mini Courses - Time Series Part 1/2 online without registration, duration hours minute second in high quality. This video was added by user George Zoto 11 November 2021, don't forget to share it with your friends and acquaintances, it has been viewed on our site 2,900 once and liked it 65 people.