GraphLab Create offers multiple ways to work with your data beyond your desktop or laptop.

  • Batch processing with job scheduling in a distributed system, using a Hadoop Yarn or EC2 cluster running Turi Distributed.

Explore the detailed functionality via the following API documentation or the Deployment chapter of the User Guide, as well as the Gallery.

Where did predictive_service go? It is now in its own package, psclient.

Distributed/Asynchronous Execution

job.create Execute arbitrary functions in a remote environment.
map_job.create Distributed execution of a function once for each entry in the parameter_set.
Job Monitor/Manage code executing on remote environments.

Ec2 Cluster

Ec2Config Config used for starting up EC2 host(s)
ec2_cluster.create Create a Turi Distributed cluster in EC2.
ec2_cluster.load Loads an EC2 Cluster object from the s3_path.
Ec2Cluster An object for managing Turi EC2 Clusters.

Hadoop Cluster

hadoop_cluster.create Create a Turi Distributed cluster in Hadoop.
HadoopCluster HadoopCluster represents a Turi Distributed cluster backed up by a Hadoop system.

Session Management

environments View and manage Environments available in the session.
jobs View and manage all Jobs in the session.


util.generate_schema Generate schema from a given sample input and output.