How do I deploy my custom model I have trained on workbench GCP?

[nitya_r]

I have trained a detectron2 model on vertex ai workbench. i have NOT used tensorflow, xgboost or scikit-learn. 

i have a model.pth file and a metrics.json file stored in my bucket when i run the model. 

How do i deploy this model on GCP and further evaluate it? Is it possible for me to create an endpoint and directly deploy my model on this or will i have to use something like cloud build?

[sascha_heyer]

Hi Nitya

To deploy your model you have to pack it into a serving container this can be either done with a custom prediction container or with custom prediction routine. 

I have written a deep dive article and video
https://medium.com/google-cloud/serving-machine-learning-models-with-google-vertex-ai-5d9644ededa3

https://www.youtube.com/watch?v=brNMT7Snlh0

Best regards