Releases · awslabs/realtime-fraud-detection-with-gnn-on-dgl · GitHub

This repository was archived by the owner on Jul 9, 2024. It is now read-only.

21 Apr 06:57

zxkane

v2.0.5 Latest

Latest

Changelogs

fix: add missing sagemaker:AddTags to sfn execute role
fix: explicitly set log bucket object ownership for S3 ACL change

Assets 2

21 Nov 02:55

zxkane

v2.0.4

pin Neptune engine version to 1.2.0.1

Assets 2

20 Nov 08:56

zxkane

v2.0.4-rc0 Pre-release

Pre-release

pin Neptune engine version to 1.2.0.1

Assets 2

28 Aug 10:51

zxkane

v2.0.3

Changelogs

fix: learning rate and epoch in training

Assets 2

18 Aug 02:06

zxkane

v2.0.2

Changelogs

feat: (experimental)SageMaker serverless Inference
fix: bump Lambda runtime to Nodejs 16.x and Python 3.9
fix: update documentdb certificate when deploying to AWS China regions

Assets 2

15 Aug 15:45

zxkane

v2.0.2-rc0 Pre-release

Pre-release

fix: bump Lambda runtime to Nodejs 16.x and Python 3.9
feat: (experimental)SageMaker serverless Inference

Assets 2

28 Oct 08:52

zxkane

v2.0.1

Fix CloudFormation deployment due to expired Api key of graphql #272

Assets 2

28 Oct 07:47

zxkane

v2.0.1-rc0-rel Pre-release

Pre-release

Merge tag 'v2.0.1-rc0' into cfn-rel

Assets 2

28 Oct 07:45

zxkane

v2.0.1-rc0 Pre-release

Pre-release

chore: bump projen and version to 2.0.1

Assets 2

13 Sep 07:28

zxkane

v2.0.0

It’s an end-to-end solution for real-time fraud detection which leverages graph database Amazon Neptune, Amazon SageMaker and Deep Graph Library (DGL) to construct a heterogeneous graph from tabular data and train a Graph Neural Network(GNN) model to detect fraudulent transactions in the IEEE-CIS dataset.

model training pipeline
online business monitor
online docs

Assets 2