tfma.default_extractors

View source on GitHub

Returns the default extractors for use in ExtractAndEvaluate.

eval_shared_model Shared model (single-model evaluation) or list of shared models (multi-model evaluation). Required unless the predictions are provided alongside of the features (i.e. model-agnostic evaluations).
eval_config Eval config.
slice_spec Deprecated (use EvalConfig).
materialize True to have extractors create materialized output.
enable_batched_extractors True if batched extractors should be used.
tensor_adapter_config Tensor adapter config which specifies how to obtain tensors from the Arrow RecordBatch. If None, we feed the raw examples to the model.
custom_predict_extractor Optional custom predict extractor for non-TF models.

NotImplementedError If eval_config contains mixed serving and eval models.