Image and content moderation services.

There are 5 available for API

• Closed questions(Standard Criteria (5 mins response time)) - Answer can be only approved, declined or ban(kenta).
• Choices(Yes or No Question from Image (30 mins response time)) - This model is used to ask a question with multiple choice. Anwser can be one or multiple.
• Photo tags(Tag an object in the image (60 mins response time)) - This model is used to create a selection area to find where the answer is, by dragging the area on an image from webpage.
• Messages(Message Question from Image (30 mins response time)) - This model allows the moderator to type the answer on what they see.
• Predictions(Images (AI Beta / 95% accuracy)) - Use AI to predict the result

Please see more usage documentation for details on our guidelines.

Getting Started

KSequencing 0.1.25 works with Rails 4.1 onwards. You can add it to your Gemfile with:

gem 'k_sequencing', '~> 0.1.25'

Then run bundle install

Next, you need to run the generator:

$ rails generate k_sequencing:install

The generator will install an initializer which describes ALL of KSequencing's configuration options. It is imperative that you take a look at it. When you are done, you are ready to add KSqeuencing to any of your codes.

You have to contact us Datawow to get your token. With this token you'll be able to use our gem properly.

Usage

There are 3 operations of each model as the same name, see example below

Get image

KSequencing.client.find_image(id)

Field	Type	Required	Description
token	string	Yes	Project token
id	string	Yes	Image id or Client's image id

Note:

• You must choose id or custom_id for search. Not both.
• Image data dynamic by project token.

Sample request

KSequencing.client.find_image("5a40be59fb9d7f27354c5efa")

or

KSequencing.client.find_image("your custom id")

or

KSequencing.client.find_image("5a40be59fb9d7f27354c5efa", { token: "[you_token]" })

Client can check whether request is successful by

response = KSequencing.client.find_image("5a40be59fb9d7f27354c5efa")
if response.successful?
  # Do stuff
else
  log.error("Request was not successful, something went wrong.")
end

Sample response

<KSequencing::Response @status=200, @message="success" @meta={"code"=>200, "message"=>"success"}, @data={}, />

{
  "data": {
    "image": {
      "id": "5a40be59fb9d7f27354c5efa",
      "answer": "approved",
      "credit_charged": 1,
      "custom_id": "custom_id",
      "data": "image_url",
      "postback_url": "postback_url",
      "processed_at": "2017-12-25T16:02:00.599+07:00",
      "project_id": "project_id",
      "status": "processed"
    }
  },
  "status": 200,
  "message": "success",
  "meta": {
    "code": 200,
    "message": "success"
  }
}

---

Create Image Standard Criteria (5 mins response time)

Closed questions - Answer can be only approved, declined or ban(kenta).

KSequencing.image_closed_question.create()

Field	Type	Required	Description
token	string	Yes	Project token
data	string	Yes	Data for moderation
postback_url	string	No	Image postback url
postback_method	string	No	Postback method
custom_id	string	No	Custom id

Sample request

KSequencing.image_closed_question.create({ data: "image_url" })

or

Override data [:token, :postback_url, :postback_method]

KSequencing.image_closed_question.create({
  custom_id: "custom_id",
  data: "image_url",
  postback_method: "POST",
  postback_url: "https://example.com/callbacks",
  token: "[your_token]"
})

Sample response

{
  "data": {
    "id": "5a40c77ffb9d7f27354c60c2",
    "answer": nil,
    "credit_charged": 0,
    "custom_id": "custom_id",
    "data": "image_url",
    "postback_url": "https://example.com/callbacks",
    "processed_at": nil,
    "project_id": "project_id",
    "status": "unprocess" 
  },
  "status": 200,
  "message": "success",
  "meta": { 
    "code": 200, 
    "message": "success" 
  }
}

Sample postback data

POST "https://example.com/callbacks?answer=declined&custom_id=custom_id&image_id=5a40cfc2fb9d7f27354c62b5&task_id=5a40cfc2fb9d7f27354c62b5"

---

Get list of images

KSequencing.image_closed_question.all({ token: "[your_token]" })

Field	Type	Required	Description
token	string	Yes	Project token
page	integer	No	Image id
per_page	integer	No	Client's image id

Note: You must choose id or custom_id for search. Not both.

Sample request

KSequencing.image_closed_question.all

or

KSequencing.image_closed_question.all({
  page: 1,
  per_page: 20,
  token: "[your_token]"
})

Sample results

{
  "data": {
    "images": [
      {
        "answer": "approved",
        "credit_charged": 1,
        "custom_id": "custom_id",
        "data": "image_url",
        "id": "5a40c77ffb9d7f27354c60c2",
        "postback_url": "https://example.com/callbacks",
        "processed_at": "2017-12-25T16:40:19.699+07:00",
        "project_id": "project_id",
        "status": "processed"
      },
      {
        "answer": "approved",
        "credit_charged": 1,
        "custom_id": "custom_id",
        "data": "image_url",
        "id": "5a40be59fb9d7f27354c5efa",
        "postback_url": "https://example.com/callbacks",
        "processed_at": "2017-12-25T16:02:00.599+07:00",
        "project_id": "project_id",
        "status": "processed"
      },
      ...
    ]
  },
  "status": 200,
  "message": "success",
  "total": 3,
  "meta": {
    "code": 200,
    "current_page": 1,
    "message": "success",
    "next_page": 2,
    "prev_page": -1,
    "total_count": 3,
    "total_pages": 2
  }
}

However token: "[your_token]" you can configure in configuration file, but it is not necessary to send to the method every time you request, if you have one project token we recommend this approach

Please see more usage documentation for details on our guidelines.

License

This project is licensed under datawowio