gcp-ml-model

Deploying machine learning model on GCP

GCP Setup

step 1

Enable Billing

step 2

Go to IAM Admin > Managed Resources > Create New Project

step 3

Go to App Engine > Dashboard > Select the project

step 4

Download the Google Cloud SDK

step 5

Open terminal and login into the cloud SDK

---

IDE

step 1

[main].py for GCP

Method 3

Deploy a model to an endpoint

1. Endpoint : Location in which to run predictions
2. Model : Container to use (ModelContainerSpec)
3. DeployedModel : Compute resources to use for online prediction

https://us-central1-aiplatform.googleapis.com/v1/projects/{project}/locations/{location}/endpoints/{endpoint}:predict

References

1. https://medium.com/analytics-vidhya/google-cloud-platform-custom-model-upload-rest-api-inference-and-model-version-monitoring-80216e69fbc2
2. Vertex AI - https://cloud.google.com/vertex-ai/docs/samples/aiplatform-upload-model-sample#aiplatform_upload_model_sample-python
3. Deploy a model to an endpoint : https://cloud.google.com/vertex-ai/docs/general/deployment