I have lexicon dataset of around 12000 words. Which has 3-4 sub categories.
What I want to do is train BERT model on my dataset. I would like to see confusion matrix and accuracy metrics to understand model performance.
After that, I want to test this model on some different records I will provide.