In the current POC that I have made, I update the Firebase MLKit Custom Model uploaded on the Firebase console manually.
So now every time I train the model with new data, I have to go manually upload the newer model tflite
file on the console.
Is there a way to automate this? A Server-To-Server API like we have for FCM?
I know it has been a while since you asked this question, but Firebase ML-Kit has recently released an API to manage your custom models. You can use the Firebase Admin SDKs (Python/Node.js) to achieve this. See: https://firebase.google.com/docs/ml-kit/manage-hosted-models
I hope this helps you and anyone else looking for a similar solution.