EdgeClient.inferences.build_inference_request

Prepare input data into inference request object for Modzy Inference API

EdgeClient.inferences.build_inference_request(model_identifier: str, model_version: str, input_sources: List[InputSource], explain=False, tags=None)

This method allows users to prepare input data into the expected InferenceRequest object format.

Parameters

Parameter	Type	Description	Example
`model_identifier`	`str`	The model identifier.	`'ed542963de'`
`model_version`	`str`	The model version string in semantic version format.	`'1.0.1'`
`input_sources`	`List[InputSource]`	A list of input sources of type `InputSource`	`[InputSource(key="input.txt", text="Today is a great day.")]`
`explain`	`bool`	If the model supports explainability, flag this job to return an explanation of the predictions	`True`
`tags`	`Mapping[str, str]`	An arbitrary set of key/value tags to associate with this inference.

Returns

A InferenceRequest object that can be submitted to Inference API

Examples

import os
from modzy import EdgeClient
from modzy.edge import InputSource

# generate requests iterator to pass to stream method
requests = []
for img in os.listdir(img_folder):
  input_object = InputSource(
    key="image", # input filename defined by model author
    data=open(os.path.join(img_folder, img), 'rb').read()
  )
  with EdgeClient('localhost', 55000) as client:
    requests.append(client.inferences.build_inference_request("<model-id>", "<model-version>", input_object, explain=False, tags=None))

Updated 9 months ago