HPE Machine Learning Development Environment CLI#

Reference

HPE Machine Learning Development Environment CLI Reference

To use HPE Machine Learning Development Environment, you’ll need, at minimum, the HPE Machine Learning Development Environment command-line interface (HPE Machine Learning Development Environment CLI) and an HPE Machine Learning Development Environment cluster. The HPE Machine Learning Development Environment CLI includes the det command-line tools for interacting with an HPE Machine Learning Development Environment cluster. This page contains instructions for using the CLI, including installion and upgrade.

Warning

Although HPE Machine Learning Development Environment supports password-based authentication, communication between the HPE Machine Learning Development Environment CLI, HPE Machine Learning Development Environment WebUI, and HPE Machine Learning Development Environment master does not take place over an encrypted channel by default.

Note

All users should install the HPE Machine Learning Development Environment CLI on their local development machine.

Note

You can also interact with HPE Machine Learning Development Environment using the web interface (WebUI).

Installation#

The CLI is distributed as a Python wheel package and requires Python >= 3.7. We recommend setting up a virtualenv and using the pip utility to install determined into the environment:

pip install determined

Note

The command, pip install determined, installs the determined library which includes the Determined command-line interface (CLI).

After installing the CLI, configure it to connect to the HPE Machine Learning Development Environment master at the appropriate IP address. To do this, set the DET_MASTER environment variable:

export DET_MASTER=<master IP>

Place this into the appropriate configuration file for your login shell, such as .bashrc.

Environment Variables#

  • DET_MASTER: The network address of the master of the HPE Machine Learning Development Environment installation. The value can be overridden using the -m flag.

  • DET_USER and DET_PASS: Specifies the current HPE Machine Learning Development Environment user and password for use when non-interactive behaviour is required such as scripts. det user login is preferred for normal usage. Both DET_USER and DET_PASS must be set together to take effect. These variables can be overridden by using the -u flag.

Verifying Installation#

To verify that the HPE Machine Learning Development Environment CLI has been installed correctly, use the following command:

det --version

This command displays the installed version of the HPE Machine Learning Development Environment CLI. If the installation was successful, you should see the version number in the output.

Uninstalling#

If you need to uninstall the HPE Machine Learning Development Environment CLI, use the following command:

pip uninstall determined

This command uninstalls the determined library, including the HPE Machine Learning Development Environment CLI, from your system.

Upgrading#

To upgrade the HPE Machine Learning Development Environment CLI to the latest version, use the following command:

pip install --upgrade determined

This command upgrades determined (along with the HPE Machine Learning Development Environment CLI) to the latest available version.

Getting Help#

Using the -h or --help argument on objects or actions prints a help message and exits the CLI. For example, to print usage for the deploy command, run the following:

det deploy -h

Similarly, you can get help for a subcommand. For example, to get help for deploy aws:

det deploy aws -h

Getting Started#

After installing the HPE Machine Learning Development Environment CLI, you can start using it to interact with your HPE Machine Learning Development Environment cluster. The CLI is invoked with the det command.

CLI subcommands usually follow a <noun> <verb> form, similar to the paradigm of ip. Certain abbreviations are supported, and a missing verb is the same as list, when possible. The following examples show different ways to achieve the same outcome using the full <noun> <verb> form, then with an abbreviation, and finally with an implicit list:

# List all experiments.
$ det experiment list
$ det e list
$ det e
# List all agents.
$ det agent list
$ det a list
$ det a
# List all slots.
$ det slot list
$ det slot
$ det s

Usage Examples#

Task

Example

Command

Options

List all experiments.

Display a list of all experiments in the cluster.

det experiment list

List all experiments for a specific network address.

Display a list of all experiments in the cluster at network address 1.2.3.4.

det -m 1.2.3.4 e

View a snapshot of logs.

Display the most recent logs for a specific command.

det command logs <command_id>

-f, –tail

View logs for a trial.

Show the logs for trial 289 and continue streaming logs in real-time.

det t logs -f 289

-f

Add a label.

Add the label foobar to experiment 17.

det e label add 17 foobar

Create an experiment.

Create an experiment in a paused state with the configuration file const.yaml and the code contained in the current directory. The paused experiment is not scheduled on the cluster until activated.

det e create -f --paused const.yaml .

Describe an experiment.

Display information about experiment 493, including full metrics, in CSV format.

det e describe 493 --metrics --csv

Set max slots.

Ensure that experiment 85 does not use more than 4 slots in the cluster.

det e set max-slots 85 4

Display details about the CLI and master.

Show detailed information about the CLI and master. This command does not take both an object and an action.

det version

Stop (kill) a command.

Terminate a running command.

det command kill <command_id>

Set a password for the admin user.

Set the password for the admin user during cluster setup.

det user change-password admin

Create a user.

Create a new user named hoid who has admin privileges.

det u create --admin hoid