Project Configuration

Contents

Projects contain the following sections:

• General
• Display
  • Inputs
  • Outputs
• Model

General

This section is the general configuration for the project.

parameter	required	description
type	true	Type of annotation task you want to complete. Please pick one of the following: classification, labeling, or translation. More options are to come.
title	true	The name of the project
slug	false	The URL slug for the project
bgColorClass	false	The background color class for the project. Please use Tailwind CSS class names
guidelines	true	The guidelines for the project.

TOML Example:

[general]
title = "Project title"
slug = "project-title"
bgColorClass = "bg-pink-600"
guidelines = """
Project guidelines go here.
"""

Display

This defines the inputs and outputs for the annotation.

Inputs

The data to display to the end-user from the dataset sample and/or predefined text.

TOML Example:

[display]
[[display.inputs]]
type = "text"
value = "Some text here"
# text provided directly to the end-user.

[[display.inputs]]
type = "text"
language = "Java"
value = "model.data.methods.*.src_code"
# text provided from the sample.

Outputs

Similar to inputs, outputs describe what form of input the user will provide as the annotation output.

parameter	required	description
type	true	Options are short-text, long-text, range, multiple-choice, or radio
range	false	A tuple defined as an array of two numbers [min, max]; Required for type range
prompt	false	An optional bit of text to describe what information to input. Markdown is accepted.
values	false	An array of strings; Required for type multiple-choice and radio

TOML Examples:

[display]
[[display.outputs]]
type = "range"
range = [ 1, 10 ]

[[display.outputs]]
type = "radio"
prompt = "![alt text](url)"
values = [ "option 1", "option 2", "option 3", "option 4" ]

Model

The model section defines the dataset sample schema. This is used for dataset validation and defining references for the inputs.

TOML Example:

[model]
type = "object"
required = true

[model.of.type]
type = "string"
required = true

[model.of._id]
type = "string"
required = true

[model.of.dataId]
type = "string"
required = true

JSON Example:

{
  "model": {
    "type": "object",
    "required": true,
    "of": {
      "type": {
        "type": "string",
        "required": true
      },
      "_id": {
        "type": "string",
        "required": true
      },
      "dataId": {
        "type": "string",
        "required": true
      }
    }
  }
}