Azure Machine Learning for the Developer
summary
Step-by-step demo of building a predictive pipeline: prepare a 10,000-row CSV in Excel/Power Query, convert profit to four labels, train multiple models in Azure Machine Learning Studio with a 70/30 train/test split, compare a decision forest (≈72% accuracy) and a neural network (≈77–78%), publish as a request‑response web service and call it from an ASP.NET app via JSON and an API key, scale to thousands of predictions per second (MS docs cite up to 5,000/s) and automate retraining.
AI-written summary · about
chapters (26)
- 0:00Intro
- 0:48Who am I
- 1:06Outline
- 2:15Good Data Science
- 2:57Predict Customer Profitability
- 9:17Data Scientist
- 10:39Challenges
- 11:49Azure Machine Learning
- 13:01How does it work
- 13:22Creating a workspace
- 14:15Moving forward
- 15:49Demo
- 18:29Adventureworks
- 18:42Preparing Data
- 25:52Creating Experiments
- 30:16Developing Models
- 31:11Uploading Data
- 32:16Data Visualization
- 33:04Column Selector
- 34:24Algorithms
- 36:41Training and Testing
- 41:09Confusion Matrix
- 42:46Publishing
- 46:18Prediction
- 48:52Deployment
- 51:48Integration Demo
related talks
59mPredicting the future as a service with Azure ML and R
58mDeep Dive into Azure Automated Machine Learning
41mLeveraging Automated Machine Learning to enable anyone to develop Machine Learning solutions
48mMachine Learning in Azure Databricks
51mTaking Models to the Next Level with Azure Machine Learning
37mModel Creation in Azure AutoML and Ingestion Through Excel
53mImplementing AI using Azure Machine Learning Services
01h 01mAI and ML imbedded in Azure Data Platform