| This model is part of the Research topic "Bias and Fairness in AI" conducted by Shaina Raza, Deepak John Reji, Chen Ding | |
| - language: en | |
| - tags: Token Classification | |
| - Dataset : MBAD Data | |
| - datasize : 17775 entries | |
| - Label distribution : Biased - 10651, Non-Biased - 7124 | |
| - Train-Test split : 90 : 10 | |
| - tokenizer : distilbert-base-uncased | |
| - model : distilbert-base-uncased | |
| - optimizer : adam | |
| - Learning rate : 5e-5 | |
| - Model parameters : 66,955,010 | |
| - epochs : 30 | |
| - Train accuracy : 0.7697 | |
| - Test accuracy : 0.62 | |
| - Train loss : 0.4506 | |
| - Test loss : 0.9644 | |
| - Carbon emission 0.319355 Kg | |