Missed TensorFlow Dev Summit? Check out the video playlist. Watch recordings

tfx.components.infra_validator.request_builder.build_requests

View source on GitHub

Build a list of request protos to be queried against the model server.

tfx.components.infra_validator.request_builder.build_requests(
    model_name, examples, request_spec
)

Args:

  • model_name: Name of the model. For example, tensorflow SavedModel is saved under directory {model_name}/{version}. The same directory structure is reused in a tensorflow serving, and you need to specify model_name in the request to access it.
  • examples: An Examples artifact which contains gzipped TFRecord file containing tf.train.Example.
  • request_spec: A RequestSpec config.

Returns:

A list of request protos.