Response Format Reference
Hive serves three response formats to support its main groups of models:
- Classification — For Visual and Text classification models.
- Detection — For Visual Detection models.
- Audio — For Speech interpretation models.
The API returns model results in the output object of the JSON response. Within each model type, the output object shares the same format.
This is an example of the returned API response.
{
"id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
"project_id": 0,
"status": [
{
"status": {
"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": [
... <Model results will be in this section, and format differs based on the model used!>
]
}
]
}
}
],
"error": "string",
"from_cache": true,
"metadata": "string"
}
The actual model response is returned inside status.response.output.