PyCM Report

Dataset Type :

Note 1 : Recommended statistics for this type of classification highlighted in aqua

Note 2 : The recommender system assumes that the input is the result of classification over the whole data rather than just a part of it. If the confusion matrix is the result of test data classification, the recommendation is not valid.

Confusion Matrix :

Actual Predict
0 1 2
0 13 0 0
1 0 15 1
2 0 3 6

Overall Statistics :

95% CI (0.79716,0.99231)
ACC Macro 0.92982
ARI 0.73605
AUNP 0.91458
AUNU 0.90555
Bangdiwala B 0.82692
Bennett S 0.84211
CBA 0.83333
CSI 0.76488
Chi-Squared 53.85218
Chi-Squared DF 4
Conditional Entropy 0.35951
Cramer V 0.84177
Cross Entropy 1.56133
F1 Macro 0.87745
F1 Micro 0.89474
FNR Macro 0.13194
FNR Micro 0.10526
FPR Macro 0.05695
FPR Micro 0.05263
Gwet AC1 0.84537
Hamming Loss 0.10526
Joint Entropy 1.90651
KL Divergence 0.01432
Kappa 0.8355
Kappa 95% CI (0.68301,0.98799)
Kappa No Prevalence 0.78947
Kappa Standard Error 0.0778
Kappa Unbiased 0.83514
Krippendorff Alpha 0.83731
Lambda A 0.81818
Lambda B 0.8
Mutual Information 1.13011
NIR 0.42105
NPV Macro 0.95108
NPV Micro 0.94737
Overall ACC 0.89474
Overall CEN 0.17658
Overall J (2.38947,0.79649)
Overall MCC 0.83929
Overall MCEN 0.24726
Overall RACC 0.36011
Overall RACCU 0.3615
P-Value 0.0
PPV Macro 0.89683
PPV Micro 0.89474
Pearson C 0.7657
Phi-Squared 1.41716
RCI 0.73051
RR 12.66667
Reference Entropy 1.54701
Response Entropy 1.48962
SOA1(Landis & Koch) Almost Perfect
SOA2(Fleiss) Excellent
SOA3(Altman) Very Good
SOA4(Cicchetti) Excellent
SOA5(Cramer) Very Strong
SOA6(Matthews) Strong
SOA7(Lambda A) Very Strong
SOA8(Lambda B) Very Strong
SOA9(Krippendorff Alpha) High
SOA10(Pearson C) Strong
Scott PI 0.83514
Standard Error 0.04978
TNR Macro 0.94305
TNR Micro 0.94737
TPR Macro 0.86806
TPR Micro 0.89474
Zero-one Loss 4

Class Statistics :

Class 0 1 2 Description
ACC 1.0 0.89474 0.89474 Accuracy
AGF 1.0 0.92297 0.799 Adjusted F-score
AGM 1.0 0.88655 0.87294 Adjusted geometric mean
AM 0 2 -2 Difference between automatic and manual classification
AUC 1.0 0.90057 0.81609 Area under the ROC curve
AUCI Excellent Excellent Very Good AUC value interpretation
AUPR 1.0 0.88542 0.7619 Area under the PR curve
BB 1.0 0.83333 0.66667 Braun-Blanquet similarity
BCD 0.0 0.02632 0.02632 Bray-Curtis dissimilarity
BM 1.0 0.80114 0.63218 Informedness or bookmaker informedness
CEN 0 0.22934 0.35141 Confusion entropy
DOR None 95.0 56.0 Diagnostic odds ratio
DP None 1.09038 0.96383 Discriminant power
DPI None Limited Poor Discriminant power interpretation
ERR 0.0 0.10526 0.10526 Error rate
F0.5 1.0 0.85227 0.81081 F0.5 score
F1 1.0 0.88235 0.75 F1 score - harmonic mean of precision and sensitivity
F2 1.0 0.91463 0.69767 F2 score
FDR 0.0 0.16667 0.14286 False discovery rate
FN 0 1 3 False negative/miss/type 2 error
FNR 0.0 0.0625 0.33333 Miss rate or false negative rate
FOR 0.0 0.05 0.09677 False omission rate
FP 0 3 1 False positive/type 1 error/false alarm
FPR 0.0 0.13636 0.03448 Fall-out or false positive rate
G 1.0 0.88388 0.75593 G-measure geometric mean of precision and sensitivity
GI 1.0 0.80114 0.63218 Gini index
GM 1.0 0.89981 0.8023 G-mean geometric mean of specificity and sensitivity
HD 0 4 4 Hamming distance
IBA 1.0 0.86946 0.45131 Index of balanced accuracy
ICSI 1.0 0.77083 0.52381 Individual classification success index
IS 1.54749 0.98489 1.85561 Information score
J 1.0 0.78947 0.6 Jaccard index
LS 2.92308 1.97917 3.61905 Lift score
MCC 1.0 0.79218 0.69332 Matthews correlation coefficient
MCCI Very Strong Strong Moderate Matthews correlation coefficient interpretation
MCEN 0 0.32202 0.42664 Modified confusion entropy
MK 1.0 0.78333 0.76037 Markedness
N 25 22 29 Condition negative
NLR 0.0 0.07237 0.34524 Negative likelihood ratio
NLRI Good Good Poor Negative likelihood ratio interpretation
NPV 1.0 0.95 0.90323 Negative predictive value
OC 1.0 0.9375 0.85714 Overlap coefficient
OOC 1.0 0.88388 0.75593 Otsuka-Ochiai coefficient
OP 1.0 0.85373 0.71164 Optimized precision
P 13 16 9 Condition positive or support
PLR None 6.875 19.33333 Positive likelihood ratio
PLRI None Fair Good Positive likelihood ratio interpretation
POP 38 38 38 Population
PPV 1.0 0.83333 0.85714 Precision or positive predictive value
PRE 0.34211 0.42105 0.23684 Prevalence
Q None 0.97917 0.96491 Yule Q - coefficient of colligation
QI None Strong Strong Yule Q interpretation
RACC 0.11704 0.19945 0.04363 Random accuracy
RACCU 0.11704 0.20014 0.04432 Random accuracy unbiased
TN 25 19 28 True negative/correct rejection
TNR 1.0 0.86364 0.96552 Specificity or true negative rate
TON 25 20 31 Test outcome negative
TOP 13 18 7 Test outcome positive
TP 13 15 6 True positive/hit
TPR 1.0 0.9375 0.66667 Sensitivity, recall, hit rate, or true positive rate
Y 1.0 0.80114 0.63218 Youden index
dInd 0.0 0.15 0.33511 Distance index
sInd 1.0 0.89393 0.76304 Similarity index

Generated By PyCM Version 4.0