Deploying sample MNIST model with TensorFlow Serve
train.py
: Trains the model and exports it in SavedModel
format.batching.config
: Contains dynamic batching configuration.requirements.txt
: Contains the dependencies.SavedModel
format checkpoint is self contained and can be deployed without writing any extra code.
Log the Model To Model Registry
Push the code to a Git repository or directly deploy from local machine
MODEL_DIR
environment variable where the model will be downloaded to.Download Model from Model Registry in the deployment configuration
MODEL_DIR
environment variable to the deployed service.Add the model id and revision from HuggingFace Hub in Artifacts Download
sectionView the deployment, logs and metrics