From f57a3efdbd20e68087f0f59e39548212c3a0dfa6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Aur=C3=A9lien=20Geron?= Date: Sun, 27 Oct 2019 11:52:13 -0700 Subject: [PATCH] Add documentation for the `mode` parameter --- g3doc/api_docs/python/tfma/load_eval_results.md | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/g3doc/api_docs/python/tfma/load_eval_results.md b/g3doc/api_docs/python/tfma/load_eval_results.md index abbadd6e59..11db9143a3 100644 --- a/g3doc/api_docs/python/tfma/load_eval_results.md +++ b/g3doc/api_docs/python/tfma/load_eval_results.md @@ -23,7 +23,12 @@ Run model analysis for a single model on multiple data sets. * `output_paths`: A list of output paths of completed tfma runs. * `mode`: The mode of the evaluation. Currently, tfma.DATA_CENTRIC_MODE - and tfma.MODEL_CENTRIC_MODE are supported. + and tfma.MODEL_CENTRIC_MODE are supported. In tfma.MODEL_CENTRIC_MODE, the + data will be presented in a model centric way which shows how a model + progresses over each run (i.e., the main axis will be the model run id). In + DATA_CENTRIC_MODE, the data will be presented in a data centric way which + shows how the latest version of the model performs as new evaluation data + become available (i.e., the main axis is the last data span). #### Returns: