At the core of Rezon is a command-line interface (CLI) that offers explainability to AI models. It allows users to easily deploy AI models, request explanations for their predictions, and generate reports on model usage. This document provides an overview of the available commands and their usage.
init command creates a new
rezon.toml configuration file with placeholder values in the current working directory.
This will create a
rezon.toml file with the following content:
[config] api_key = "your_api_key_here" [model] h5_file = "/path/to/h5_file_model1.h5" json_path = "/path/to/json_file_model.json" parameters = ["status", "credit duration", "credit history", "credit purpose", "credit amount", "savings", "employment duration", "installment rate", "personal status", "other debtors", "residence length", "property", "age", "other installment plans", "housing", "number credits", "job quality", "number of people liable", "telephone", "foreign worker"]
deploy command deploys an AI model by looking for a rezon.toml configuration file in the current directory.
This will start the process of deploying your model to Rezon cloud:
run command sends input parameters to your hosted model and retrieves the actual perdiction and explanation.
rezon run [OPTIONS] --input "[<input_values>...]"
Options: -i, --input [<input_values>...] A 1D or 2D JSON array of input values -f, --input_file [<input_file>...] Path to CSV file with input values -e, --explain Include explanations for your model's output -j, --json Run your model and receive the result as json -h, --help Print help
report command generates a report of all the specified model's explanations during a past time period
rezon report [OPTIONS]
Options: -d, --days <days> Specify the number of days to generate the report for. Defaults to the last 30 days. -i, --instances <instances> Specify the number of instances to generate the report for. Specify either "instances" or "days", not both. -h, --help Print help