This model was trained using the 'bert-base-uncased' from the transformer library and it was trained on the popular fake/real news dataset from Kaggle. | |
Pytorch is the framework used to train the model and it had an accuracy score of 93.5 % and here is what the classification report looks like. | |
precision recall f1-score support | |
0 0.96 0.92 0.94 2348 | |
1 0.92 0.96 0.94 2142 | |
accuracy 0.94 4490 | |
macro avg 0.94 0.94 0.94 4490 | |
weighted avg 0.94 0.94 0.94 4490 | |
--- | |
license: apache-2.0 | |
--- | |