Learn about aspects of Machine Learning that require intuition, good judgment and experimentation to finely tune and optimize your ML models for the best performance.
Learn the many knobs and levers involved in training a model. Manually adjust them to see their effects on model performance. Once familiar with the hyperparameters, you will learn how to tune them in an automatic way using Cloud Machine Learning Engine on Google Cloud Platform.
Course Objectives:
- Generalize your model
- Tune batch size and learning rate for better model performance
- Optimize your model
- Differentiate between parameters and hyperparameters
- Think beyond simple grid search algorithms
- Take advantage of Cloud ML Engine for hyperparameter tuning
- Apply L1 and L2 regularization for sparsity
- Understand why regularization is important for Logistic Regression
- How to train Neural Networks effectively
- How to work with multi-class Neural Networks
- Use embeddings to: Manage sparse data, Reduce dimensionality, Increase model generalization, Cluster observations
- Create reusable embeddings
- Explore embeddings in TensorBoard
- Go beyond canned estimators, write a custom estimator
- Incorporate Keras models into Estimator
Labs and Demos: Lab: Training Data Analyst
Lab: Improve model accuracy by Hyperparameter Tuning with Cloud AI Platform
Lab: Practicing with L1 Regularization