Deploy pretrained model

Takes in a TFLite file and builds the model and SDK. Updates are streamed over the websocket API (or can be retrieved through the /stdout endpoint). Use getProfileTfliteJobResult to get the results when the job is completed.

POST /api/{projectId}/jobs/deploy-pretrained-model

Try it

Deploy pretrained model

cURL

curl --request POST \
  --url https://studio.edgeimpulse.com/v1/api/{projectId}/jobs/deploy-pretrained-model \
  --header 'Content-Type: application/json' \
  --header 'x-api-key: <api-key>' \
  --data '
{
  "modelFileBase64": "<string>",
  "modelFileType": "tflite",
  "deploymentType": "<string>",
  "modelInfo": {
    "input": {
      "inputType": "time-series",
      "frequencyHz": 123,
      "windowLengthMs": 123
    },
    "model": {
      "modelType": "classification",
      "labels": [
        "<string>"
      ]
    }
  },
  "engine": "tflite",
  "representativeFeaturesBase64": "<string>",
  "deployModelType": "int8",
  "useConverter": "onnx-tf"
}
'

200

{
  "success": true,
  "id": 12873488112,
  "error": "<string>"
}

Authorizations

x-api-key · string · header · required

Path Parameters

projectId · integer · required

Project ID

Query Parameters

impulseId · integer

Impulse ID. If this is unset then the default impulse is used.

Body

application/json

modelFileBase64 · string · required

A base64 encoded pretrained model

modelFileType · enum<string> · required

Available options: tflite, onnx, saved_model, lgbm, xgboost, pickle

deploymentType · string · required

The name of the built target. You can find this by listing all deployment targets through listDeploymentTargetsForProject (via GET /v1/api/{projectId}/deployment/targets) and see the format type.

modelInfo · object · required

Show child attributes

modelInfo. input · object · required

Option 1

Show child attributes

modelInfo.input. inputType · enum<string> · required

Available options: time-series

modelInfo.input. frequencyHz · number · required

modelInfo.input. windowLengthMs · integer · required

modelInfo. model · object · required

Option 1

Show child attributes

modelInfo.model. modelType · enum<string> · required

Available options: classification

modelInfo.model. labels · string[] · required

engine · enum<string>

Available options: tflite, tflite-eon, tflite-eon-ram-optimized, tensorrt, tensaiflow, drp-ai, tidl, akida, syntiant, memryx, neox, ethos-linux, st-aton, ceva-npn

representativeFeaturesBase64 · string

A base64 encoded .npy file containing the features from your validation set (optional for onnx and saved_model) - used to quantize your model.

deployModelType · enum<string>

Available options: int8, float32

useConverter · enum<string>

Optional, use a specific converter (only for ONNX models).

Available options: onnx-tf, onnx2tf

Response

200 - application/json

OK

success · boolean · required

Whether the operation succeeded

id · integer · required

Job identifier. Status updates will include this identifier.

Example:

12873488112

error · string

Optional error description (set if 'success' was false)