MultiFusionBERT / _report.txt
pierpaologo's picture
Latest version with results
6fbd10c verified
******************** CLASSIFICATION REPORT ALL BIO LABELS ********************
precision recall f1-score support
B-AdHominem 0.68 0.48 0.57 27
B-AppealtoAuthority 0.71 0.57 0.63 30
B-AppealtoEmotion 0.64 0.54 0.59 120
B-FalseCause 0.86 0.67 0.75 9
B-Slipperyslope 0.71 0.62 0.67 8
B-Slogans 0.00 0.00 0.00 5
I-AdHominem 0.85 0.91 0.88 712
I-AppealtoAuthority 0.86 0.80 0.83 1019
I-AppealtoEmotion 0.73 0.83 0.78 2104
I-FalseCause 0.91 0.88 0.89 312
I-Slipperyslope 0.91 0.91 0.91 324
I-Slogans 0.00 0.00 0.00 44
O 0.94 0.92 0.93 7914
accuracy 0.89 12628
macro avg 0.68 0.63 0.65 12628
weighted avg 0.89 0.89 0.88 12628
******************** CLASSIFICATION REPORT ALL NORMALIZED LABELS ********************
precision recall f1-score support
AdHominem 0.85 0.90 0.88 739
AppealtoAuthority 0.86 0.80 0.83 1049
AppealtoEmotion 0.74 0.83 0.79 2224
FalseCause 0.91 0.87 0.89 321
O 0.94 0.92 0.93 7914
Slipperyslope 0.90 0.91 0.91 332
Slogans 0.00 0.00 0.00 49
accuracy 0.89 12628
macro avg 0.75 0.75 0.75 12628
weighted avg 0.89 0.89 0.89 12628
******************** CLASSIFICATION REPORT BINARY LABELS ********************
precision recall f1-score support
Not fallacy 0.94 0.92 0.93 7914
Fallacy 0.87 0.90 0.89 4714
accuracy 0.91 12628
macro avg 0.91 0.91 0.91 12628
weighted avg 0.92 0.91 0.91 12628