Model based trained on 30% of the kaggle public data on fake and reals news article. The model achieved an `auc` of 1.0, precision, recall and f1score all at score of 1.0.

* Task;- The predictor classifies news articles into either fake or real news. 
* It is a transformer model trained using the `ktrain` library on 30% of dataset of size 194MB after preprocessing.
* Metrics used are recall,, precision, f1score and roc_auc_score.