From the course: Data Science on Google Cloud Platform: Predictive Analytics

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Jobs

Jobs

- [Instructor] Training models usually run for a long amount of time. In Cloud ML, you can manage them using Jobs. Jobs are long-running operations that can either be used for training models or for batch predictions. They are asynchronous in nature. They can either be created through the command line or through the REST API. You can set up training jobs that can consume data from Cloud Storage and then store the models built back into Cloud Storage. You can run prediction batch jobs that consume models and data from Cloud Storage and write the predictions back into Cloud Storage. Cloud ML provides user interfaces through which you can track the status of a given job. It also provides access to log files generated by your code. This helps in troubleshooting any issues that are found.

Contents