i172-python-invasion-claim-identify

We trained three model to identify claim for python invasion.

Least XLnet and bert_multi_label were trained on Colab. The other one was trained with Comprehend on AWS.

Training code for Bert and XLnet are in 172_claim_prediction_bert_multi and 172_claim_prediction_xlnet_single notebooks.

XLnet data and trained model can be found at: https://drive.google.com/drive/u/0/folders/1Iy-TqgGbU6odNKgj4EIIqdPpW3Zk-Xb-

Bert data and trained model can be found at: https://drive.google.com/drive/u/0/folders/1MTUngI1CusJmrAGrICsj5-n6MWpOHxRW

Model summary

XLnet confusion matrix

Bert multi label confusion matrix

AWS Comprehend confusion matrix