AD Release Note
v2.0.3
June. 13. 2024
Bug Fixes Fix stopping training pipeline error
- Deactivate graph display option
Compatibility: ALO 2.5.0 Tested on c6i.2xlarge - CPU: 6 / MEM: 14G, p3.2xlarge - CPU: 6 / MEM: 55G / GPU: 1(16G) / CUDA 11
v2.0.2
May. 07. 2024
Improvements Modify inference_summary.yaml
Compatibility: ALO 2.3.4 Tested on c6i.2xlarge - CPU: 6 / MEM: 14G, p3.2xlarge - CPU: 6 / MEM: 55G / GPU: 1(16G) / CUDA 11
v2.0.1
May. 07. 2024
Improvements Update for Mellerikat requirements
- Support ALO v2.3.4
Compatibility: ALO 2.3.4 Tested on c6i.2xlarge - CPU: 6 / MEM: 14G, p3.2xlarge - CPU: 6 / MEM: 55G / GPU: 1(16G) / CUDA 11
v2.0.0
Mar. 28. 2024
Improvements Refactor code architecture
- Enhanced user convenience by integrating the machine learning anomaly detection pipeline, allowing users to focus on individual contents when they want to perform anomaly detection (Integration with CAD planned for the future).
- Restructured common functions used in both DT and SR algorithms into the StatisticModel class for easier code management.
- Modified variable names that were difficult to understand.
Reduce computational cost
- Reduced running time of average calculation function within the DT algorithm.
- Reduced memory usage by removing unnecessary code declarations within the model.
Compatibility: ALO 2.0