How nantiny99/wafer-level-anomaly-detection-with-mlflow works
This repository is a notebook-based wafer anomaly-detection experiment. The documented workflow evaluates Isolation Forest and Autoencoder models on two ST-AWFD wafer datasets, compares results with imbalance-aware metrics, and tracks runs and artifacts in MLflow. The tree contains only the notebook and README, and no source excerpts were provided, so the graph reflects documented capabilities rather than inspected notebook internals.
Main entry points: Detecting Anomaly with MLflow.ipynb Experiment Notebook
Components 11
Wafer Data
- ST-AWFD Dataset
- Wafer 1 and 2
Detecting Anomaly with MLflow.ipynb
Model Experiments
- Experiment Notebook
Detecting Anomaly with MLflow.ipynb - Isolation Forest
Detecting Anomaly with MLflow.ipynb - Autoencoder
Detecting Anomaly with MLflow.ipynb - Threshold Selection
Detecting Anomaly with MLflow.ipynb
Evaluation and Tracking
- Metric Evaluation
Detecting Anomaly with MLflow.ipynb - MLflow Tracking
- Model Artifacts
Detecting Anomaly with MLflow.ipynb - Performance Comparison
Detecting Anomaly with MLflow.ipynb
Other components
- Experimenter
Connections 11
- Experimenter → Experiment Notebook: runs experiments
README.md - ST-AWFD Dataset → Wafer 1 and 2: provides data
README.md - Experiment Notebook → Wafer 1 and 2: evaluates datasets
README.md - Experiment Notebook → Isolation Forest: evaluates model
README.md - Experiment Notebook → Autoencoder: evaluates model
README.md - Autoencoder → Threshold Selection: requires tuning
README.md - Experiment Notebook → Metric Evaluation: assesses performance
README.md - Experiment Notebook → MLflow Tracking: tracks runs
README.md - MLflow Tracking → Model Artifacts: tracks artifacts
README.md - Experiment Notebook → Performance Comparison: compares results
README.md - Metric Evaluation → Performance Comparison: informs comparison
README.md