Data Science on Google Cloud Platform: Predictive Analytics

Learn how to use Google Cloud Platform to train and deploy machine learning models for predictive analytics.

Introduction

• Why use predictive analytics on GCP?
• Data science modules covered

1. ML Options in GCP

• Cloud Dataproc
• Cloud ML Engine
• Cloud Natural Language
• Cloud Translation
• Cloud Vision
• Cloud Video Intelligence
• Cloud Dialogflow

2. Cloud ML Basics

• Models
• Model versions
• Jobs
• Predictive analytics process

3. Model Building with Cloud ML

• Understanding input data
• Build and test model locally
• Upload files to Cloud Storage
• Modify code to work with GCP
• Creating a training package
• Running training synchronously
• Training using jobs

4. Predictions in Cloud ML

• Creating a deployment model
• Creating a model version
• Creating a prediction dataset
• Running a prediction

5. Cloud ML Best Practices

• Cost control
• Local testing
• Performance monitoring

Conclusion

• Next steps