Figure Eight Manage Job Data API
Load data to jobs and work with that data.
Load data to jobs and work with that data.
openapi: 3.0.0
info:
title: Appen Platform Account Info Manage Job Data API
version: 1.0.0
description: "# Introduction\nHello, and welcome to Appen, the leading data annotation platform for Machine Learning. Our API enables developers to build applications that can programmatically create, edit, and launch Appen jobs, download results, and more! \n\nAppen uses a RESTful API that accepts data requests as URL-encoded key-value pairs. Responses are returned in JSON and authentication is key-based.\n\nBefore integrating with the API, we highly recommend you first build and run a job using our [graphical user interface](https://client.appen.com/sessions/new). Once you have run a job successfully and are satisfied with the results, you can automate the process using the Appen API.\n\n# Authentication\nYou will need to supply your Appen API key to access the API. To find your API key:\n1. Click the Account button located in the bottom corner of the Global Navigation bar on the left side of any page.\n\n2. Click the API tab and find your key listed under the Your API Key section. \n\n\nFor requests listed below, the API Key will be listed as `{api_key}`. \n\nThe API key needs to be included in the header for every request: \n```\n headers: { \n\n \"AUTHORIZATION\" => \"Token token={api_key}\" \n\n } \n```\n\nYou may also see other variables, such as `{job_id}` for operations pertaining to a specific job, or `{team_id}` when completing operations relevant to an entire team (like listing all available jobs). Job ID's can be determined from the job listing page or the URL. A team id is visible in the URL of the Teams tab in the Account page. \n# Resources\nThere are three primary data objects on Appen that correspond to resources on the API:\n## Jobs\nA **Job** is an interface that connects your data to an online workforce.\n Each job on the Appen platform has data, instructions, customizable questions for your use case\n (written in **CML** [](https://success.appen.com/hc/en-us/articles/202817989-CML-Custom-Markup-Language-Overview)),\n [test questions](https://success.appen.com/hc/en-us/articles/202702985-How-to-Create-Test-Questions),\n and is worked on by *Contributors*. Contributors submit *Judgments* on the rows of data via a worker interface. \n * All jobs in a single account can be found [on your jobs page](https://client.appen.com/jobs) and will be\n identified by a unique numeric id.\n## Units (also known as 'Rows')\nA **Unit** of data is uploaded to the job from your source data file via API or GUI. \n## Judgments \nA **Judgment** is the set of responses submitted by a contributor on a particular unit of data. It is\n recommended to collect multiple judgments and compare them to one another or aggregate to the top response. For\n each job, you can specify the number of judgments you would like each row to receive.\n# Responses and Messaging\nEach request to the Appen API returns an HTTP status code response and one or both of the following:\n* An application response message (in JSON) \n* A JSON representation of the object (job, unit, or judgments) requested\nSee the below sections of different API Requests, Responses and Messaging for more information:\n1. [Job Create/Update](/#tag/Job-CreateUpdate)\n2. [Manage Job Data](/#tag/Manage-Job-Data)\n3. [Job Status](/#tag/Job-Status)\n4. [Monitor Contributors](/#tag/Monitor-Contributors)\n5. [Manage Job Settings](/#tag/Manage-Job-Settings)\n6. [Job Results](/#tag/Job-Results)\n7. [Account Info](/#tag/Account-Info)\n"
license:
name: Licensed under Apache 2.0
url: http://www.apache.org/licenses/LICENSE-2.0.html
termsOfService: https://www.appen.com/privacy-statement/
contact:
email: help@appen.com
servers:
- url: https://api.appen.com/v1
description: Production Server
tags:
- name: Manage Job Data
description: Load data to jobs and work with that data.
externalDocs:
url: https://success.appen.com/hc/en-us/articles/202703435-Jobs-Resource-Attributes
paths:
/jobs/{job_id}/upload.json:
put:
tags:
- Manage Job Data
summary: Upload data to an existing job
description: Upload data to an existing job. Data may be uploaded in JSON or CSV format.
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
- in: query
name: force
description: 'The request fails when the fields in the data uploaded _do not
match_ the fields already present in the job. To override this
validation and force upload data, use the query parameter
`force=true`.
'
required: false
schema:
type: boolean
requestBody:
description: 'Accepts JSON or CSV. This action corresponds to uploading a CSV using
the "Add More Data" button displayed in the UI. Note that this request
invokes the upload operation. CSV files must be UTF-8 encoded.
'
content:
multipart/form-data:
schema:
type: string
format: binary
responses:
'200':
description: success
'404':
$ref: '#/components/responses/404'
x-code-samples:
- lang: Shell
label: cURL - CSV
source: "curl -X PUT \\\n\"https://api.appen.com/v1/jobs/{job_id}/upload.json\" \\\n-H \"Authorization: Token token={api_key}\" \\\n-H \"Content-Type: text/csv\" \\\n-T \"./additional_data.csv\" \n"
- lang: Shell
label: cURL - JSON
source: 'curl -X PUT \
"https://api.appen.com/v1/jobs/{job_id}/upload.json" \
-H "Authorization: Token token={api_key}" \
-H "Content-Type: application/json" \
-T "./additional_data.json"
'
/jobs/{job_id}/units.json:
get:
tags:
- Manage Job Data
summary: Paginated list of Units in Job
description: 'Gets **Units** for the **Job** identified by the `job_id` parameter of
the request. Defaults to 1000 Units per page, in order to get the next
1000 rows, you must increment the value of `page`. Keep doing so until
an empty hash is returned.
'
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
- $ref: '#/components/parameters/page'
responses:
'200':
description: Displays 1000 units at a time.
content:
application/json:
schema:
type: array
items:
type: string
example:
- id: 1234,
- data: {}
x-code-samples:
- lang: Shell
label: cURL
source: 'curl -X GET \
"https://api.appen.com/v1/jobs/{job_id}/units.json" \
-H "Authorization: Token token={api_key}" \
-d "page=1"
'
post:
tags:
- Manage Job Data
summary: Add Unit(s) to job
description: 'Creates a new Row in an existing Job. This is the request format and endpoint to use for unit-by-unit posting. This endpoint cannot add multiple units in a single request.
*Note:* test questions can also be uploaded using this endpoint. First, upload units that have a field `_golden=true`, then use the [Convert uploaded test questions](#tag/Manage-Job-Data/paths/~1jobs~1{job_id}~1gold.json/put) endpoint.
'
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
- in: query
name: unit[data][{column1,2,3...x}]
description: Accepts multiple columns, and individual data strings per column
required: false
schema:
type: string
- in: query
name: unit[data][column1,2,...x_gold]
description: Accepts multiple columns, and individual data strings per column
required: false
schema:
type: string
- in: query
name: unit[data][_golden]
description: TBD
required: false
schema:
type: boolean
responses:
'201':
description: 'JSON object of the row created, including its id, data, and related
Job id. JSON application message indicating status of the operation
(e.g., Unit was successfully created)
'
content:
application/json:
schema:
$ref: '#/components/schemas/unit'
x-code-samples:
- lang: Shell
label: cURL
source: 'curl -X POST \
"https://api.appen.com/v1/jobs/{job_id}/units.json" \
-d "unit[data][image_url]=some_data_1" \
-d "unit[data][description]=some_data_2" \
-d "unit[data][uid]=some_data_3" \
-H "Authorization: Token token={api_key}" \
'
/jobs/{job_id}/units/{unit_id}.json:
put:
tags:
- Manage Job Data
summary: Change unit state.
description: Update `state` for specific units on a job.
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
- $ref: '#/components/parameters/unit_id'
- in: query
name: unit[state]
description: 'Changes the state (e.g., new, golden, finalized, canceled,
hidden_gold) of the Row specified in the request
'
required: false
schema:
type: string
responses:
'200':
description: 'JSON object comprising Judgment count, Row state, and timestamp values.
'
content:
application/json:
schema:
type: array
items:
$ref: '#/components/schemas/unit'
x-code-samples:
- lang: Shell
label: cURL
source: "curl -X PUT \\\n\"https://api.appen.com/v1/jobs/{job_id}/units/{unit_id}.json\" \\\n-H \"Authorization: Token token={api_key}\" \\\n--data-urlencode \"unit[state]={new}\" \n"
delete:
tags:
- Manage Job Data
summary: Delete a unit
description: Delete a unit if the job state and unit state are valid for deletion.
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
- $ref: '#/components/parameters/unit_id'
responses:
'200':
description: OK
content:
application/json:
schema:
type: object
properties:
message:
type: object
properties:
success:
type: string
example: Unit {unit_id} has been deleted.
x-code-samples:
- lang: Shell
label: cURL
source: 'curl -X DELETE \
"https://api.appen.com/v1/jobs/{job_id}/units/{unit_id}.json" \
-H "Authorization: Token token={api_key}" \
'
/jobs/{job_id}/units/{unit_id}/cancel.json:
post:
tags:
- Manage Job Data
summary: Cancel a unit.
description: 'Will cancel a unit in state New or Judgable if the row has not collected any judgments yet.
'
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
- $ref: '#/components/parameters/unit_id'
responses:
'200':
description: OK
content:
application/json:
schema:
type: object
properties:
success:
type: string
example: Unit {unit_id} has been canceled.
x-code-samples:
- lang: Shell
label: cURL
source: 'curl -X POST \
"https://api.appen.com/v1/jobs/{job_id}/units/{unit_id}/cancel.json" \
-H "Authorization: Token token={api_key}" \
'
/jobs/{job_id}/gold.json:
put:
tags:
- Manage Job Data
summary: Convert uploaded test questions.
description: 'Corresponds to the "Convert Uploaded Test Questions" button on the Job''s DATA page.
All units where `_golden` is set to `true` will be converted to test questions.
'
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
responses:
'200':
description: Test Question Rows in the specified Job
content:
application/json:
schema:
$ref: '#/components/schemas/unit'
x-code-samples:
- lang: Shell
label: cURL
source: 'curl -X PUT \
"https://api.appen.com/v1/jobs/{job_id}/gold.json" \
-H "Authorization: Token token={api_key}" \
'
/jobs/{job_id}/test_questions:
get:
tags:
- Manage Job Data
summary: Get all test question units.
description: Get all test question units.
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
responses:
'200':
description: Test Question Rows in the specified Job
content:
application/json:
schema:
$ref: '#/components/schemas/unit'
x-code-samples:
- lang: Shell
label: cURL
source: 'curl -X GET \
"https://api.appen.com/v1/jobs/{job_id}/gold.json" \
-H "Authorization: Token token={api_key}" \
'
/jobs/{job_id}/units/ping.json:
get:
tags:
- Manage Job Data
summary: Get count of all rows in a job.
description: Get count of all rows in a job.
parameters:
- $ref: '#/components/parameters/job_id'
- $ref: '#/components/parameters/api_key'
responses:
'200':
description: JSON object containing count of Rows in Job
x-code-samples:
- lang: Shell
label: cURL
source: "curl -X GET \\\n\"https://api.appen.com/v1/jobs/{job_id}/units/ping.json\" \\\n-H \"Authorization: Token token={api_key}\" \n"
components:
schemas:
new_job_id:
description: Integer identifier of the requested resource (Job, Unit, Workflow) used in API request paths and response bodies.
type: integer
format: int64
readOnly: true
minimum: 1
maximum: 9223372036854776000
example: 1234567
unit:
properties:
id:
type: integer
format: int64
readOnly: true
example: 1234
data:
type: object
example: {}
judgments_count:
type: integer
format: int32
state:
type: string
enum:
- new
- judgable
- judging
- judged
- ordering
- golden
- hidden_gold
- canceled
- finalized
- preprocessing
agreement:
type: number
multipleOf: 0.01
missed_count:
type: integer
format: int32
contested_count:
type: integer
format: int32
gold_pool:
type: string
created_at:
type: string
format: date-time
readOnly: true
updated_at:
type: string
format: date-time
readOnly: true
job_id:
type: integer
format: int32
readOnly: true
example: 425765
parameters:
page:
in: query
name: page
description: "A desired page of paginated results to return. \n"
required: false
schema:
type: integer
api_key:
name: key
in: query
required: true
description: "Your personal Appen API key; used to access the Appen platform. Can be found in your [account settings](https://client.appen.com/account/api). Variable name: `{api_key}`. \n"
schema:
type: string
unit_id:
name: unit_id
in: path
required: true
description: "A unique identifier for a row of data that has been loaded into the system. Variable name: `{unit_id}`. \n"
schema:
type: integer
format: int32
job_id:
in: path
name: job_id
description: "A unique identifier for a job or task that is distributed to contributors. Variable name: `{job_id}`. \n"
required: true
schema:
$ref: '#/components/schemas/new_job_id'
responses:
'404':
description: 'Not Found. The resource you are referencing does not exist, or you don''''t own it.
'