- A guide to building an interactive flexdashboard
- Visualization for machine learning algorithms:
- By Chris Fonnesbeck
- Hyperparameter optimization for Neural Networks
- Kaggle ensembling guide
- All the little of visualization design
- Case Study:
Owing a lot of thanks to MOOC, I took on the challenge as a full-time MOOC Data Engineer/Analyst at the Centre for Teaching, Learning and Technology at UBC. My manager, Ido Roll, embarked on the new venture and started the project.
UBC has been hosting dozens of courses on edX, with more than 150,000 registered students. We built and maintained the data pipeline that on a daily base, transformed data in its various formats, e.g. SQL dump, JSON, XML, CSV to structured data and stored them in Google BigQuery. I generated customized data reports and performed data analysis for research purpose. A major part of my work is to streamline and automate the dashboard reporting process. Our final product is VizIT, an interactive dashboard that seeks to offer an intuitive understanding of students’ learning and provide success metrics and actionable information for the course. VizIT is quite a pioneer in this area.
Feel free to check out my Github repository for some of my work.
- Implementation of a basic reproducible data analysis workflow
- The reason why I’m not the biggest fan of Kaggle – The “Perfect Score”: a way to hack the LeaderBoard in Kaggle
- A Kaggler’s Guide to Model Stacking in Practice
- A Kaggle Master Explains Gradient Boosting
- 21 Must-Know Data Science Interview Questions and Answers
- How do you know if your model is going to work?
- Case Study: