Microaggression Detection Model - Training Configuration
============================================================

Model Name: microsoft/deberta-base
Max Length: 128
Batch Size: 32
Epochs: 5
Learning Rate: 2e-05
Patience: 2
Dataset Shape: (52275, 10)
Training Samples: 41820
Validation Samples: 10455
Model Parameters: 139,193,858

NOTE: Model trained on combined text + context using [SEP] token