Post Wml Instances Instance Published Models Published Model Deployments
POST {{baseUrl}}/wml_instances/:instance_id/published_models/:published_model_id/deployments?sync=<>
Creates the deployment - online, stream, batch (batch deployment only supports Cloud Object Storage as input/output for Tensorflow, Keras and Caffe models.)
Request Params
| Key | Datatype | Required | Description |
|---|---|---|---|
sync | string | This is parameter which determines if the 'online' deployment needs to be created in synchronous or asynchronous way. The 'batch' and 'stream' deployments can be created only in synchronous way. In case the online deployment is not ready within 90 seconds the synchronous request will automatically switch to asynchronous mode and return 202 Accepted |