Data Science Fundamentals
summary
Covers practical steps from business goal to production: follow CRISP-DM style iterations, build a one-row-per-outcome dataset, do feature engineering, split train/test to detect overfitting, use squared-loss awareness to handle outliers, and set up monitoring and regular retraining. Lists concrete tech: Azure event hubs/queues for buffering, Databricks + Spark for pipelines, Delta Lake for streaming CRUD on features, scoring APIs or containers, and model deployment options including ONNX and Custom Vision.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 0:38About Steph
- 1:57What is Data Science
- 2:14Models
- 5:41Business Value
- 6:41Data Mining Process
- 14:23Skills Needed
- 15:21Functions Needed
- 20:45How to Succeed
- 21:37First Project
- 23:17Preparing for Production
- 27:14Linear Regression
- 29:17Bias Variance
- 30:36Loss minimization
- 34:20Outliers
- 37:14Polynomials
- 41:47RealTime Prediction
- 43:46Results
- 44:48Java
- 45:38Scoring API
- 46:03Microsoft Stack
- 50:25Data Lakes
related talks
47mA Practical(ish) Introduction to Data Science
36mCore of Data Science and Open Source Resources
06mInterview
38mGetting ready for a career in Analytics and Data Science
04h 02mHands-on Session on Data Science for Beginners
01h 26mData Driven Action: a primer on Data Science
44mPeople, Technology, Culture: How to make Data Science work
24mData Science: the best path to align business objectives