Transformer Classifier

Transformer encoder-only model using pytorch is used to classify signal and background events with a neural network attached to it, with the low level particle information as input features. The evaluation is done with k=2 fold cross validation. Latest training performance are shown in the following plots.