>>> Hands-On Walkthroughs¶
- Introduction
- Class diagrams and inheritance charts of Neuraxle objects
- The Mixin design pattern in machine learning
- Steps containing other steps as the composite design pattern in machine learning
- Scikit-learn’s pipeline.Pipeline class and how to shift to parallel deep learning
- FeatureUnion to compute steps in parallel and join their results
- AutoML module to automatically tune hyperparameters of your pipelines
- All the base classes of Neuraxle together
- Automatic Hyperparameter Tuning / AutoML
- AutoML loop
- Define your problem
- Define your pipeline
- Choose a validation splitter
- Define a the main scoring metric with a first MetricsCallback
- Add other metric callbacks with MetricCallback (optional)
- Select an hyperparams repository
- Select an hyperparams optimizer
- Create, and launch AutoML loop
- Get best model and measure test accuracy
- Additional note : model selection as an hyperparameter
- AutoML loop
- Handler Methods
- Step Saving & Lifecycle
- Time Series Processing Example
- Random Distributions
- REST API Serving