A classifier classifies an input (such as an entire image or sentence of text) into different categories. It assigns a confidence score for each class. Hive’s classifier models include the visual classification models and text classification models.

Classification models can be multi-headed, where each group of mutually exclusive model classes belong to a single model head. For example, when an image is run through Hive's visual moderation model, one head might classify sexually not-safe-for-work (NSFW) content while another head might classify the presence of guns.

This concept is illustrated below. This imaginary model has two heads:

NSFW classification: general_nsfw, general_suggestive, general_not_nsfw_not_suggestive
Gun classification: gun_in_hand, animated_gun, gun_not_in_hand, no_gun

The confidence scores for each model head sum to 1.

Sample of task.status.output

{
    "id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
    "code": 200,
    "project_id": 12345,
    "user_id": 1234,
    "created_on": "2020-05-25T19:55:02.312Z",
    "status": {
        "status": {
            "code": 0,
            "message": "SUCCESS"
        },
        "response": {
            "input": {
                "id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
                "charge": 0,
                "model": "string",
                "model_version": 0,
                "model_type": "string",
                "created_on": "string",
                "media": {
                    "type": "string",
                    "mimetype": "string",
                    "duration": 0,
                    "width": 0,
                    "height": 0
                },
                "user_id": 0,
                "project_id": 0
            },
            "output": [
                {
                    "time": 0,
                    "classes": [
                        {
                            "class": "general_nsfw",
                            "score": 0.9
                        },
                        {
                            "class": "general_suggestive",
                            "score": 0.05
                        },
                        {
                            "class": "general_not_nsfw_not_suggestive",
                            "score": 0.05
                        },
                        {
                            "class": "gun_in_hand",
                            "score": 0.88
                        },
                        {
                            "class": "animated_gun",
                            "score": 0.04
                        },
                        {
                            "class": "gun_not_in_hand",
                            "score": 0.04
                        },
                        {
                            "class": "no_gun",
                            "score": 0.04
                        }
                    ]
                }
            ]
        }
    },
    "from_cache": false,
    "metadata": "string"
}

Name	Description
time	Timestamp in seconds of video or audio frame extracted from the original media. Always 0 for images
classes	List of objects for each output class that the API predicts.
class	Name of predicted class.
score	Confidence score of predicted class.
id	Unique identifier for the job or request.
code	HTTP or request-level status code. In the sample, 200 indicates a successful request.
project_id	The top-level project ID associated with this request.
user_id	The top-level user ID associated with this request.
created_on	ISO 8601 timestamp indicating when this request was created.
status.status.message	Descriptive message corresponding to the status code. (e.g. 0 for success)
status.response.output	Array of output objects, each containing classification data (time and classes).
from_cache	Boolean indicating whether the result was retrieved from a cache.
metadata	Additional metadata or notes about the request.