Overview
Last updated
Last updated
Jobs have been replaced by a more general framework called Workflows. For the latest info go to Workflows.
Gradient Jobs execute generic tasks on remote infrastructure and can be used to perform a variety of functions from compiling a model to running an ETL operation. When you run an Experiment within Gradient, a job is created and executed (or multiple jobs in the case of distributed training or a hyperparameter sweep).
Gradient Jobs can be created using the Web UI or via the CLI, and they form part of a larger suite of tools that work seamlessly with Gradient Notebooks and other features of the Gradient platform.
a collection of files (code, resources, etc) from your local computer or GitHub
a container (with code dependencies and packages pre-installed)
a command to execute (i.e. python main.py or nvidia-smi)
There are several ways to run a Job in Gradient: You can use the Job Builder, a web interface for submitting Jobs, the Gradient CLI, or you can clone a public Job.
There are many features you will want to check out like outputting your model to the /artifacts directory, the persistent data layer at /storage, graphing with Job Metrics, sharing Jobs with the public Jobs feature, and more.
Jobs have access to your persistent storage (/storage) and can output files to /artifacts. Learn more about Gradient storage here.
Jobs can be chosen to run on a variety of hardware. Pricing and details for all available options can be found here.
