Paulie-Aditya/sign-language-detection

Novel Approach 1

Stacked Classifier: RF + SVM + XGB

metrics:

• —Accuracy: 0.9911734164070612
• —Balanced Accuracy: 0.9903422714760236
• —MCC: 0.990784932183338
• —ROC AUC Score: 0.999934898058849
• —F1 Score: 0.9911734164070612
• —Jaccard Score: 0.9825012866700978
• —Log Loss: 0.033553756349283356
• —Precision: 0.9911734164070612
• —Recall: 0.9911734164070612

Novel Approach 2

Stacked Classifier: RF + SVM + KNN + XGB

metrics:

• —Accuracy: 0.9922118380062306
• —Balanced Accuracy: 0.9913200369813552
• —MCC: 0.9918690348004674
• —ROC AUC Score: 0.9999193482927975
• —F1 Score: 0.9922118380062306
• —Jaccard Score: 0.9845440494590417
• —Log Loss: 0.03136301122428542
• —Precision: 0.9922118380062306
• —Recall: 0.9922118380062306