From 9162950bb5227036b2f40b972f462095ef1b79b5 Mon Sep 17 00:00:00 2001 From: Raj Sinha Date: Tue, 21 May 2024 08:31:39 -0700 Subject: [PATCH] Update README to clarify that partial matching is not implemented in the code. PiperOrigin-RevId: 635816253 --- README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.md b/README.md index 75032a2..8dc7fba 100644 --- a/README.md +++ b/README.md @@ -15,6 +15,8 @@ [Semi-supervised Pseudo Labeler Anomaly Detection with Ensembling (SPADE)](https://openreview.net/forum?id=JwDpZSv3yz) is a semi-supervised anomaly detection method that uses an ensemble of one class classifiers as the pseudo-labelers and supervised classifiers to achieve state of the art results especially on datasets with distribution mismatch between labeled and unlabeled samples. Provided with a min-max scaled dataset, and label values denoting unlabeled, positive, and negative data points located in BigQuery, this custom model trains and uploads TensorFlow saved model assets to a specified GCS location. +==Partial matching as described in the paper is not implemented in the open-source version of SPADE== + ## Assumptions - You have a Google Cloud account and are familiar with the console