BERT-GRADER?

BERT-Grader is a BERT model designed for grading text by mesuring gramar, consistency and quality. Due to an error in the dataset, 'undefined' is 0. BERT-Grader is a finetune of distil-bert, for performance.

Classes

0 (Undefined) - Elementary 1 - Intermidiate 2 - Advanced

Biases and Issues

For BERT-Grader to perform well, the input text should be an actual essay. Some biases may arise due to training data phrases, but there were mesures in place to limit bias. Giberrish is accounted for, and will be rather random. I suggest using a spell-checker with this model, and mesure tokens.

Validation Metrics

loss: 0.166 accuracy: 0.933333..

Citing

@misc {intone_2023,
    author       = { {intone} },
    title        = { BERT-GRADER (Revision 6bddc72) },
    year         = 2023,
    url          = { https://huggingface.co/intone/BERT-GRADER },
    doi          = { 10.57967/hf/1257 },
    publisher    = { Hugging Face }
}