Wrapping-up

Machine Learning cho Dữ liệu Chuỗi thời gian bằng Python

Chris Holdgraf

Fellow, Berkeley Institute for Data Science

Timeseries and machine learning

• The many applications of time series + machine learning
• Always visualize your data first
• The scikit-learn API standardizes this process

Feature extraction and classification

• Summary statistics for time series classification
• Combining multiple features into a single input matrix
• Feature extraction for time series data

Model fitting and improving data quality

• Time series features for regression
• Generating predictions over time
• Cleaning and improving time series data

Validating and assessing our model performance

• Cross-validation with time series data (don't shuffle the data!)
• Time series stationarity
• Assessing model coefficient and score stability

Advanced concepts in time series

• Advanced window functions
• Signal processing and filtering details
• Spectral analysis

Advanced machine learning

• Advanced time series feature extraction (e.g., tsfresh)
• More complex model architectures for regression and classification
• Production-ready pipelines for time series analysis

Ways to practice

• There are a lot of opportunities to practice your skills with time series data.
• Kaggle has a number of time series predictions challenges
• Quantopian is also useful for learning and using predictive models others have built.

Let's practice!

Machine Learning cho Dữ liệu Chuỗi thời gian bằng Python