Development Tips

Running locally for testing

Sometimes it is useful to be able to test your model locally without the need to build image with s2i or docker.

This can be easily done with seldon-core as its installed the CLI command that starts the microservice.

Assuming we have a simple model saved in file:

class MyModel:

    def predict(self, X, features_names=None):
        Return a prediction.

        X : array-like
        feature_names : array of feature names (optional)
        print("Predict called - will run identity function")
        return X

We can start Seldon Core microservice with

seldon-core-microservice MyModel --service-type MODEL

Then in other terminal we can send curl requests to test REST endpoint:

curl http://localhost:9000/api/v1.0/predictions \
    -H 'Content-Type: application/json' \
    -d '{"data": {"names": ["input"], "ndarray": ["data"]}}'

And assuming that seldon-core code is accessible at ${SELDON_CORE_DIR} we can use grpcurl to send gRPC request:

cd ${SELDON_CORE_DIR}/executor/proto && grpcurl \
    -d '{"data": {"names": ["input"], "ndarray": ["data"]}}' \
    -plaintext -proto ./prediction.proto seldon.protos.Seldon/Predict

The grpcurl tool can be obtained using binaries released on GitHub or using asdf-vm.

See Python Server documentation for config options.