AC-Monitor

This repository is publicly accessible, with all proprietary source files, data, and sensitive information removed.

Commit Information

The commit "Final changes" (ede9ab0) on the main branch contains the latest presentable content for the showcase.

Meaningful Artifacts

• src/notebooks/aclabeler.py: Defines the ACLabeler class.
• src/notebooks/AC_labeler.ipynb: Notebook for labeling sample data using ACLabeler from src/notebooks/aclabeler.py.
• notebooks/BART-ecg-anomaly-detection.ipynb: Anomaly detection with autoencoder and LSTM. Also available as ac_anomaly_detection.py and ac_anomaly_detection.ipynb in the notebooks and scripts sub-folders during the annotation process.
• data/tempseries_labeled.zip: First 3 rows (out of 129,760) of the actual labeled time series data used in BART-ecg-anomaly-detection.ipynb.
• data/labeledtimewindows/labeledtimewindows_ab_yuan.zip: First five rows of the labels done by Steven Yuan.
• data/labeledtimewindows/labeledtimewindows_william.zip: First five rows of the labels done by William Fei.
• data/labeledtimewindows/labeled_timewindows_conflict_yuanwilliam.zip: First five rows of the labels done by Floyd, showing conflicts.
• data/labeledtimewindows_all.zip: The combination of labels from Steven, William, and Floyd.

Achievements

• Runner-up for Data Science Lifecycle Ribbon of Excellence at Berkeley's Fall 2022 Research Showcase
• Learn more about the Fall 2022 Showcase.