Skip to main content
Join the official 2020 Python Developers SurveyStart the survey!

Arkindex CLI client easy and sexy to use

Project description

Arkindex CLI

The Arkindex CLI allows you to perform various advanced actions on an Arkindex instance. It can both be used interactively or for scripting.

You can install this tool using pip: pip install arkindex-cli

To get general help about the CLI from the command line, use arkindex -h. To get specific help for a subcommand, use arkindex <subcommand> -h.

Logging in

To interact with an Arkindex instance, you first need to log in with your email and password. To do so, use this command:

arkindex login

You will be asked for the instance URL, your email and your password. If it all goes well, you will be asked for an alias under which the credentials should be stored, and whether or not these should be the default credentials to use for all other commands.

The credentials are then stored in a YAML file at ~/.config/arkindex/cli.yaml. Your email and password are not directly stored; only the instance URL and an API token.

In any subcommand, you can use the -p or --profile arguments to select a profile other than your default. For example, if you are logged in to two instances using the aliases Foo and Bar, and your default instance is Foo, all arkindex commands will login to Foo by default, and you can connect to Bar using arkindex --profile Bar <subcommand>.

ML reports

Arkindex machine learning workers can return ml_report.json artifacts; JSON files that describe which elements a worker processed, along with the created elements, classifications or transcriptions and the encountered errors.

The CLI can fetch all of the ML reports for a process and provide statistics on the errors:

arkindex process report <Process ID>

A possible output might be:

11061 elements: 10575 successful, 486 with errors
    Errors by class
┃ Class       ┃ Count ┃
│ HTTPError   │   470 │
│ KeyError    │    15 │
│ ReadTimeout │     1 │

By default, this command retrieves the ML reports for the latest run of the process. If you want to use another run, you can specify its number using -r or --run:

arkindex process report <Process ID> --run 4

Output modes

A JSON mode is available with the -j or --json arguments. This will return an object with all elements from all ML reports that have at least one error.

You can also display the full error messages and tracebacks with syntax highlighting using -v or --verbose.

Process recovery

It is possible to start a new process on another process' failed elements (elements with at least one error):

arkindex process recover <Process ID>

This will retrieve the ML reports, list the failed elements, add them to your selection, then create an unconfigured process. A link will be provided to then open the Arkindex frontend, allowing you to configure and start your new process.

Since this updates your selection, if you already had selected elements, the tool will ask for your confirmation before deselecting them.

By default, this command retrieves failed elements from the ML reports for the latest run of the process. If you want to use another run, you can specify its number using -r or --run:

arkindex process recover <Process ID> --run 4

Classes management

You can build a CSV file listing all the ML classes from a corpus:

arkindex classes --init my_classes.csv <corpus_id>

The file my_classes.csv will then have two columns (ID and class name), for each class found.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for arkindex-cli, version 0.1.3
Filename, size File type Python version Upload date Hashes
Filename, size arkindex_cli-0.1.3-py3-none-any.whl (16.5 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size arkindex-cli-0.1.3.tar.gz (13.7 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page