Machine Learning Algorithms with H2o and PowerShell
summary
Uses H2O (open source) from Windows PowerShell via the REST API and Flow GUI translating UI calls into HTTP posts, with concrete demos: Iris (150 rows) classification (predict returns per-class confidence like 0.99), MNIST image digits (~20 MB dataset, training takes ~10 minutes on a laptop), k‑means clustering, grid search for hyperparameters and AutoML that runs for hours. Shows exporting a model as a POJO→WAR web service that responds to REST calls and reports performance (example: naive deep net MSE ~3.12 vs. grid best ≈0).
AI-written summary · about
chapters (29)
- 0:00Start
- 3:17Api Driven
- 4:38Gui
- 5:51Sparkling Water
- 7:42Grid Searching
- 10:15Recommendation Engines
- 11:47H2o Jar
- 16:08Parse this File
- 25:54Percentage of Confidence
- 27:18Powershell Environment
- 27:55Convert To Form Data
- 31:42Parse Setup
- 33:56Call the Rest Api for Splitting the Frames
- 37:27Training the Deep Learning Model
- 38:32Import and Render
- 41:49Import H2o Data
- 48:03Overfitting
- 50:15Decision Trees
- 50:51Multiple Decision Trees What Is a Decision Tree
- 51:49Distributed Random Forest
- 57:12Results
- 1:03:30Grid Search
- 1:05:48Bayesian Optimization
- 1:10:02Prediction Service Builder
- 1:18:01Automl
- 1:18:41Run Automl
- 1:19:38Early Stopping
- 1:22:29Resources
- 1:23:33The Stanford Coursera Course
related talks
01h 18mIntroduction to H20 AutoML with Python
44mScalable Automatic Machine Learning with H2O
01h 10mIntroduction to Machine Learning with H2O and Python
50mScalable Data Science & Deep Learning with H2O
30mScalable Automatic Machine Learning with H2O
01h 10mGetting started with H2O on Python
27mDeploying distributed AI and machine learning in financial services
43mMachine Learning with R