EdgeClient.jobs.submit_embedded
Submit a job with the Edge client using bytes or byte arrays as inputs
EdgeClient.jobs.submit_embedded(identifier, version, sources, explain=False)
Submit a job with the Edge client based on byte array inputs, the sources dictionary is a double dict
structure as follows:
{
'<<<<input-item-key>>>>': {
'<<<<data-input-item-key>>>>: b'This is binary content'
}
}
Where:
<<<<input-item-key>>>>
: are user defined keys to identify the input items.
<<<<data-input-item-key>>>>
: are model defined keys (usually file names).
Parameters
Parameter | Type | Description | Example |
---|---|---|---|
identifier | str Model | Model identifier provided by Modzy or a model object previusly loaded. | 'ed542963de' |
version | str | The model’s version number. It follows the semantic versioning format. | '1.0.1' |
sources | dict | A mapping of source names to byte arrays. Each source should be a mapping of model input key to a byte array. | '{'my-input': {'input.txt': bytearray([1,2,3,4])}}' |
explain | bool | If the model supports explainability, flag this job to return an explanation of the predictions | True |
Returns
A Job
object with the status from the server
{
"job_identifier": "string",
"status",
"account_identifier": "string",
"explain": "boolean",
"created_at": "date-time",
"updated_at": "date-time",
"submitted_at": "date-time",
"submitted_by": "string",
"pending": "integer",
"completed": "integer",
"failed": "integer",
"total": "integer",
"model": {
"identifier": "string",
"version": "string",
"name": "string"
},
"job_inputs": ["string"],
"user": {
"identifier": "string",
"external_identifier": "string",
"email": "string",
"firstName": "string",
"lastName": "string",
"status": "string",
"title": "string"
"access_keys": [
{
"prefix": "string",
"is_default": "boolean"
}
]
}
}
Examples
from modzy import EdgeClient
client = EdgeClient("localhost", 55000)
job = client.jobs.submit_embedded('model-identifier', '1.2.3',
{
'source-name-1': {
'model-input-name-1': b'some bytes',
'model-input-name-2': bytearray([1,2,3,4]),
},
'source-name-2': {
'model-input-name-1': b'some bytes',
'model-input-name-2': bytearray([1,2,3,4]),
}
})
Updated 9 months ago