CLI

The RI CLI can be run as follows:

rime-engine [COMMAND] [ARGUMENTS]

Commands

For an up-to-date list of available commands, run --help:

rime-engine --help

The valid types of commands are:

create-firewall: Create a firewall.
run-firewall: Run firewall on a batch of tabular data.
run-firewall-images: Run firewall on a batch of Image data.
run-firewall-nlp: Run firewall on a batch of NLP data.
run-images: Run RIME on image data
run-images-local: Run RIME on image data and save to local file system
run-nlp: Run RIME on NLP data
run-nlp-local: Run RIME on NLP data and save to local file system
run-stress-tests: Run RIME offline tests on tabular data
run-stress-tests-local: Run RIME offline tests on tabular data and save to local file system
upload: Upload Results

Different commands require different arguments.

The main arguments for most of these commands are:

--config-path: Path to configuration
--upload-endpoint: Endpoint to which results will be uploaded. [env var: RIME_UPLOAD_URL]
- --disable-tls (optional): Whether to disable TLS encryption, which is enabled by default. Disabling TLS is recommended for local uploads. [env var: RIME_DISABLE_TLS]
--api-key: The api key providing authentication to RI service. Required for external uploads. [env var: RIME_API_KEY]
--save-path (optional): Path to save the results to. This will store the results locally as well as upload them to the specified endpoint
--run-name (optional): Name of the Test Run. Overrides the Test Run Name in the configuration
--firewall-endpoint: Endpoint for interacting with the firewall [env var: RIME_FIREWALL_URL]
--disable-firewall-events: Disable processing events at a datapoint level when running firewall over a batch of data.

For an up-to-date list of available arguments for a given command, run --help with the command name:

rime-engine [COMMAND] --help

For stress testing only (run-stress-tests, run-nlp, and run-images):

For firewall only (run-firewall, run-firewall-nlp, and run-firewall-images):