- run main.py to create model based on magicdataset.csv
- mace_ml_model.pkl is trained model on magicdataset.csv
- scaler.pkl is a scaler model A scaler is a preprocessing tool used to normalize or standardize feature values so that all input features are on the same scale. In our case, we used StandardScaler from sklearn.preprocessing to scale the MAGIC dataset's features before training the model.
- app.py contains flask api code
The dataset consists of 10 continuous features and 1 binary class label. The features are listed below:
fLength: continuous # major axis of ellipse [mm]
fWidth: continuous # minor axis of ellipse [mm]
fSize: continuous # 10-log of sum of content of all pixels [in #phot]
fConc: continuous # ratio of sum of two highest pixels over fSize [ratio]
fConc1: continuous # ratio of highest pixel over fSize [ratio]
fAsym: continuous # distance from highest pixel to center, projected onto major axis [mm]
fM3Long: continuous # 3rd root of third moment along major axis [mm]
fM3Trans: continuous # 3rd root of third moment along minor axis [mm]
fAlpha: continuous # angle of major axis with vector to origin [deg]
fDist: continuous # distance from origin to center of ellipse [mm]
class: g,h # gamma (signal), hadron (background)