Multi-class confusion matrix library in Python

PyCM is a multi-class confusion matrix library written in Python that supports both input data vectors and direct matrix, and a proper tool for post-classification model evaluation that supports most classes and overall statistics parameters. PyCM is the swiss-army knife of confusion matrices, targeted mainly at data scientists that need a broad array of metrics for predictive models and accurate evaluation of a large variety of classifiers.

Installation

• PyCM 2.4 is the last version to support Python 2.7 & Python 3.4
• Plotting capability requires Matplotlib (>= 3.0.0) or Seaborn (>= 0.9.1)

Source code

• Download Version 3.2 or Latest Source
• Run pip install -r requirements.txt or pip3 install -r requirements.txt (Need root access)
• Run python3 setup.py install or python setup.py install (Need root access)

PyPI

• Check Python Packaging User Guide
• Run pip install pycm==3.2 or pip3 install pycm==3.2 (Need root access)

Conda

• Check Conda Managing Package
• conda install -c sepandhaghighi pycm (Need root access)

Easy install

• Run easy_install --upgrade pycm (Need root access)

MATLAB

• Download and install MATLAB (>=8.5, 64/32 bit)
• Download and install Python3.x (>=3.5, 64/32 bit)
  • Select Add to PATH option
  • Select Install pip option
• Run pip install pycm or pip3 install pycm (Need root access)
• Configure Python interpreter

>> pyversion PYTHON_EXECUTABLE_FULL_PATH

• Visit MATLAB Examples

Docker

• Run docker pull sepandhaghighi/pycm (Need root access)
• Configuration :
  • Ubuntu 16.04
  • Python 3.6

Usage

From vector

>>> from pycm import *
>>> y_actu = [2, 0, 2, 2, 0, 1, 1, 2, 2, 0, 1, 2] # or y_actu = numpy.array([2, 0, 2, 2, 0, 1, 1, 2, 2, 0, 1, 2])
>>> y_pred = [0, 0, 2, 1, 0, 2, 1, 0, 2, 0, 2, 2] # or y_pred = numpy.array([0, 0, 2, 1, 0, 2, 1, 0, 2, 0, 2, 2])
>>> cm = ConfusionMatrix(actual_vector=y_actu, predict_vector=y_pred) # Create CM From Data
>>> cm.classes
[0, 1, 2]
>>> cm.table
{0: {0: 3, 1: 0, 2: 0}, 1: {0: 0, 1: 1, 2: 2}, 2: {0: 2, 1: 1, 2: 3}}
>>> print(cm)
Predict 0 1 2
Actual
0 3 0 0
1 0 1 2
2 2 1 3
Overall Statistics :
95% CI (0.30439,0.86228)
ACC Macro 0.72222
ARI 0.09206
AUNP 0.66667
AUNU 0.69444
Bangdiwala B 0.37255
Bennett S 0.375
CBA 0.47778
CSI 0.17778
Chi-Squared 6.6
Chi-Squared DF 4
Conditional Entropy 0.95915
Cramer V 0.5244
Cross Entropy 1.59352
F1 Macro 0.56515
F1 Micro 0.58333
FNR Macro 0.38889
FNR Micro 0.41667
FPR Macro 0.22222
FPR Micro 0.20833
Gwet AC1 0.38931
Hamming Loss 0.41667
Joint Entropy 2.45915
KL Divergence 0.09352
Kappa 0.35484
Kappa 95% CI (-0.07708,0.78675)
Kappa No Prevalence 0.16667
Kappa Standard Error 0.22036
Kappa Unbiased 0.34426
Krippendorff Alpha 0.37158
Lambda A 0.16667
Lambda B 0.42857
Mutual Information 0.52421
NIR 0.5
Overall ACC 0.58333
Overall CEN 0.46381
Overall J (1.225,0.40833)
Overall MCC 0.36667
Overall MCEN 0.51894
Overall RACC 0.35417
Overall RACCU 0.36458
P-Value 0.38721
PPV Macro 0.56667
PPV Micro 0.58333
Pearson C 0.59568
Phi-Squared 0.55
RCI 0.34947
RR 4.0
Reference Entropy 1.5
Response Entropy 1.48336
SOA1(Landis & Koch) Fair
SOA2(Fleiss) Poor
SOA3(Altman) Fair
SOA4(Cicchetti) Poor
SOA5(Cramer) Relatively Strong
SOA6(Matthews) Weak
Scott PI 0.34426
Standard Error 0.14232
TNR Macro 0.77778
TNR Micro 0.79167
TPR Macro 0.61111
TPR Micro 0.58333
Zero-one Loss 5
Class Statistics :
Classes 0 1 2
ACC(Accuracy) 0.83333 0.75 0.58333
AGF(Adjusted F-score) 0.9136 0.53995 0.5516
AGM(Adjusted geometric mean) 0.83729 0.692 0.60712
AM(Difference between automatic and manual classification) 2 -1 -1
AUC(Area under the ROC curve) 0.88889 0.61111 0.58333
AUCI(AUC value interpretation) Very Good Fair Poor
AUPR(Area under the PR curve) 0.8 0.41667 0.55
BCD(Bray-Curtis dissimilarity) 0.08333 0.04167 0.04167
BM(Informedness or bookmaker informedness) 0.77778 0.22222 0.16667
CEN(Confusion entropy) 0.25 0.49658 0.60442
DOR(Diagnostic odds ratio) None 4.0 2.0
DP(Discriminant power) None 0.33193 0.16597
DPI(Discriminant power interpretation) None Poor Poor
ERR(Error rate) 0.16667 0.25 0.41667
F0.5(F0.5 score) 0.65217 0.45455 0.57692
F1(F1 score - harmonic mean of precision and sensitivity) 0.75 0.4 0.54545
F2(F2 score) 0.88235 0.35714 0.51724
FDR(False discovery rate) 0.4 0.5 0.4
FN(False negative/miss/type 2 error) 0 2 3
FNR(Miss rate or false negative rate) 0.0 0.66667 0.5
FOR(False omission rate) 0.0 0.2 0.42857
FP(False positive/type 1 error/false alarm) 2 1 2
FPR(Fall-out or false positive rate) 0.22222 0.11111 0.33333
G(G-measure geometric mean of precision and sensitivity) 0.7746 0.40825 0.54772
GI(Gini index) 0.77778 0.22222 0.16667
GM(G-mean geometric mean of specificity and sensitivity) 0.88192 0.54433 0.57735
IBA(Index of balanced accuracy) 0.95062 0.13169 0.27778
ICSI(Individual classification success index) 0.6 -0.16667 0.1
IS(Information score) 1.26303 1.0 0.26303
J(Jaccard index) 0.6 0.25 0.375
LS(Lift score) 2.4 2.0 1.2
MCC(Matthews correlation coefficient) 0.68313 0.2582 0.16903
MCCI(Matthews correlation coefficient interpretation) Moderate Negligible Negligible
MCEN(Modified confusion entropy) 0.26439 0.5 0.6875
MK(Markedness) 0.6 0.3 0.17143
N(Condition negative) 9 9 6
NLR(Negative likelihood ratio) 0.0 0.75 0.75
NLRI(Negative likelihood ratio interpretation) Good Negligible Negligible
NPV(Negative predictive value) 1.0 0.8 0.57143
OC(Overlap coefficient) 1.0 0.5 0.6
OOC(Otsuka-Ochiai coefficient) 0.7746 0.40825 0.54772
OP(Optimized precision) 0.70833 0.29545 0.44048
P(Condition positive or support) 3 3 6
PLR(Positive likelihood ratio) 4.5 3.0 1.5
PLRI(Positive likelihood ratio interpretation) Poor Poor Poor
POP(Population) 12 12 12
PPV(Precision or positive predictive value) 0.6 0.5 0.6
PRE(Prevalence) 0.25 0.25 0.5
Q(Yule Q - coefficient of colligation) None 0.6 0.33333
QI(Yule Q interpretation) None Moderate Weak
RACC(Random accuracy) 0.10417 0.04167 0.20833
RACCU(Random accuracy unbiased) 0.11111 0.0434 0.21007
TN(True negative/correct rejection) 7 8 4
TNR(Specificity or true negative rate) 0.77778 0.88889 0.66667
TON(Test outcome negative) 7 10 7
TOP(Test outcome positive) 5 2 5
TP(True positive/hit) 3 1 3
TPR(Sensitivity, recall, hit rate, or true positive rate) 1.0 0.33333 0.5
Y(Youden index) 0.77778 0.22222 0.16667
dInd(Distance index) 0.22222 0.67586 0.60093
sInd(Similarity index) 0.84287 0.52209 0.57508
>>> cm.print_matrix()
Predict 0 1 2
Actual
0 3 0 0
1 0 1 2
2 2 1 3
>>> cm.print_normalized_matrix()
Predict 0 1 2
Actual
0 1.0 0.0 0.0
1 0.0 0.33333 0.66667
2 0.33333 0.16667 0.5
>>> cm.print_matrix(one_vs_all=True,class_name=0) # One-Vs-All, new in version 1.4
Predict 0 ~
Actual
0 3 0
~ 2 7
>>> cm = ConfusionMatrix(y_actu, y_pred, classes=[1,0,2]) # classes, new in version 3.2
>>> cm.print_matrix()
Predict 1 0 2
Actual
1 1 0 2
0 0 3 0
2 1 2 3
>>> cm = ConfusionMatrix(y_actu, y_pred, classes=[1,0,4]) # classes, new in version 3.2
>>> cm.print_matrix()
Predict 1 0 4
Actual
1 1 0 0
0 0 3 0
4 0 0 0

Direct CM

>>> from pycm import *
>>> cm2 = ConfusionMatrix(matrix={"Class1": {"Class1": 1, "Class2":2}, "Class2": {"Class1": 0, "Class2": 5}}) # Create CM Directly
>>> cm2
pycm.ConfusionMatrix(classes: ['Class1', 'Class2'])
>>> print(cm2)
Predict Class1 Class2
Actual
Class1 1 2
Class2 0 5
Overall Statistics :
95% CI (0.44994,1.05006)
ACC Macro 0.75
ARI 0.17241
AUNP 0.66667
AUNU 0.66667
Bangdiwala B 0.68421
Bennett S 0.5
CBA 0.52381
CSI 0.52381
Chi-Squared 1.90476
Chi-Squared DF 1
Conditional Entropy 0.34436
Cramer V 0.48795
Cross Entropy 1.2454
F1 Macro 0.66667
F1 Micro 0.75
FNR Macro 0.33333
FNR Micro 0.25
FPR Macro 0.33333
FPR Micro 0.25
Gwet AC1 0.6
Hamming Loss 0.25
Joint Entropy 1.29879
KL Divergence 0.29097
Kappa 0.38462
Kappa 95% CI (-0.354,1.12323)
Kappa No Prevalence 0.5
Kappa Standard Error 0.37684
Kappa Unbiased 0.33333
Krippendorff Alpha 0.375
Lambda A 0.33333
Lambda B 0.0
Mutual Information 0.1992
NIR 0.625
Overall ACC 0.75
Overall CEN 0.44812
Overall J (1.04762,0.52381)
Overall MCC 0.48795
Overall MCEN 0.29904
Overall RACC 0.59375
Overall RACCU 0.625
P-Value 0.36974
PPV Macro 0.85714
PPV Micro 0.75
Pearson C 0.43853
Phi-Squared 0.2381
RCI 0.20871
RR 4.0
Reference Entropy 0.95443
Response Entropy 0.54356
SOA1(Landis & Koch) Fair
SOA2(Fleiss) Poor
SOA3(Altman) Fair
SOA4(Cicchetti) Poor
SOA5(Cramer) Relatively Strong
SOA6(Matthews) Weak
Scott PI 0.33333
Standard Error 0.15309
TNR Macro 0.66667
TNR Micro 0.75
TPR Macro 0.66667
TPR Micro 0.75
Zero-one Loss 2
Class Statistics :
Classes Class1 Class2
ACC(Accuracy) 0.75 0.75
AGF(Adjusted F-score) 0.53979 0.81325
AGM(Adjusted geometric mean) 0.73991 0.5108
AM(Difference between automatic and manual classification) -2 2
AUC(Area under the ROC curve) 0.66667 0.66667
AUCI(AUC value interpretation) Fair Fair
AUPR(Area under the PR curve) 0.66667 0.85714
BCD(Bray-Curtis dissimilarity) 0.125 0.125
BM(Informedness or bookmaker informedness) 0.33333 0.33333
CEN(Confusion entropy) 0.5 0.43083
DOR(Diagnostic odds ratio) None None
DP(Discriminant power) None None
DPI(Discriminant power interpretation) None None
ERR(Error rate) 0.25 0.25
F0.5(F0.5 score) 0.71429 0.75758
F1(F1 score - harmonic mean of precision and sensitivity) 0.5 0.83333
F2(F2 score) 0.38462 0.92593
FDR(False discovery rate) 0.0 0.28571
FN(False negative/miss/type 2 error) 2 0
FNR(Miss rate or false negative rate) 0.66667 0.0
FOR(False omission rate) 0.28571 0.0
FP(False positive/type 1 error/false alarm) 0 2
FPR(Fall-out or false positive rate) 0.0 0.66667
G(G-measure geometric mean of precision and sensitivity) 0.57735 0.84515
GI(Gini index) 0.33333 0.33333
GM(G-mean geometric mean of specificity and sensitivity) 0.57735 0.57735
IBA(Index of balanced accuracy) 0.11111 0.55556
ICSI(Individual classification success index) 0.33333 0.71429
IS(Information score) 1.41504 0.19265
J(Jaccard index) 0.33333 0.71429
LS(Lift score) 2.66667 1.14286
MCC(Matthews correlation coefficient) 0.48795 0.48795
MCCI(Matthews correlation coefficient interpretation) Weak Weak
MCEN(Modified confusion entropy) 0.38998 0.51639
MK(Markedness) 0.71429 0.71429
N(Condition negative) 5 3
NLR(Negative likelihood ratio) 0.66667 0.0
NLRI(Negative likelihood ratio interpretation) Negligible Good
NPV(Negative predictive value) 0.71429 1.0
OC(Overlap coefficient) 1.0 1.0
OOC(Otsuka-Ochiai coefficient) 0.57735 0.84515
OP(Optimized precision) 0.25 0.25
P(Condition positive or support) 3 5
PLR(Positive likelihood ratio) None 1.5
PLRI(Positive likelihood ratio interpretation) None Poor
POP(Population) 8 8
PPV(Precision or positive predictive value) 1.0 0.71429
PRE(Prevalence) 0.375 0.625
Q(Yule Q - coefficient of colligation) None None
QI(Yule Q interpretation) None None
RACC(Random accuracy) 0.04688 0.54688
RACCU(Random accuracy unbiased) 0.0625 0.5625
TN(True negative/correct rejection) 5 1
TNR(Specificity or true negative rate) 1.0 0.33333
TON(Test outcome negative) 7 1
TOP(Test outcome positive) 1 7
TP(True positive/hit) 1 5
TPR(Sensitivity, recall, hit rate, or true positive rate) 0.33333 1.0
Y(Youden index) 0.33333 0.33333
dInd(Distance index) 0.66667 0.66667
sInd(Similarity index) 0.5286 0.5286
>>> cm2.stat(summary=True)
Overall Statistics :
ACC Macro 0.75
F1 Macro 0.66667
FPR Macro 0.33333
Kappa 0.38462
Overall ACC 0.75
PPV Macro 0.85714
SOA1(Landis & Koch) Fair
TPR Macro 0.66667
Zero-one Loss 2
Class Statistics :
Classes Class1 Class2
ACC(Accuracy) 0.75 0.75
AUC(Area under the ROC curve) 0.66667 0.66667
AUCI(AUC value interpretation) Fair Fair
F1(F1 score - harmonic mean of precision and sensitivity) 0.5 0.83333
FN(False negative/miss/type 2 error) 2 0
FP(False positive/type 1 error/false alarm) 0 2
FPR(Fall-out or false positive rate) 0.0 0.66667
N(Condition negative) 5 3
P(Condition positive or support) 3 5
POP(Population) 8 8
PPV(Precision or positive predictive value) 1.0 0.71429
TN(True negative/correct rejection) 5 1
TON(Test outcome negative) 7 1
TOP(Test outcome positive) 1 7
TP(True positive/hit) 1 5
TPR(Sensitivity, recall, hit rate, or true positive rate) 0.33333 1.0
>>> cm3 = ConfusionMatrix(matrix={"Class1": {"Class1": 1, "Class2":0}, "Class2": {"Class1": 2, "Class2": 5}},transpose=True) # Transpose Matrix
>>> cm3.print_matrix()
Predict Class1 Class2
Actual
Class1 1 2
Class2 0 5

• matrix() and normalized_matrix() renamed to print_matrix() and print_normalized_matrix() in version 1.5

Activation threshold

threshold is added in version 0.9 for real value prediction.

For more information visit Example3

Load from file

file is added in version 0.9.5 in order to load saved confusion matrix with .obj format generated by save_obj method.

For more information visit Example4

Sample weights

sample_weight is added in version 1.2

For more information visit Example5

Transpose

transpose is added in version 1.2 in order to transpose input matrix (only in Direct CM mode)

Relabel

relabel method is added in version 1.5 in order to change ConfusionMatrix classnames.

>>> cm.relabel(mapping={0:"L1",1:"L2",2:"L3"})
>>> cm
pycm.ConfusionMatrix(classes: ['L1', 'L2', 'L3'])

Position

position method is added in version 2.8 in order to find the indexes of observations in predict_vector which made TP, TN, FP, FN.

>>> cm.position()
{0: {'FN': [], 'FP': [0, 7], 'TP': [1, 4, 9], 'TN': [2, 3, 5, 6, 8, 10, 11]}, 1: {'FN': [5, 10], 'FP': [3], 'TP': [6], 'TN': [0, 1, 2, 4, 7, 8, 9, 11]}, 2: {'FN': [0, 3, 7], 'FP': [5, 10], 'TP': [2, 8, 11], 'TN': [1, 4, 6, 9]}}

To array

to_array method is added in version 2.9 in order to returns the confusion matrix in the form of a NumPy array. This can be helpful to apply different operations over the confusion matrix for different purposes such as aggregation, normalization, and combination.

>>> cm.to_array()
array([[3, 0, 0],
 [0, 1, 2],
 [2, 1, 3]])
>>> cm.to_array(normalized=True)
array([[1. , 0. , 0. ],
 [0. , 0.33333, 0.66667],
 [0.33333, 0.16667, 0.5 ]])
>>> cm.to_array(normalized=True,one_vs_all=True, class_name="L1")
array([[1. , 0. ],
 [0.22222, 0.77778]])

Combine

combine method is added in version 3.0 in order to merge two confusion matrices. This option will be useful in mini-batch learning.

>>> cm_combined = cm2.combine(cm3)
>>> cm_combined.print_matrix()
Predict Class1 Class2
Actual
Class1 2 4
Class2 0 10

Plot

plot method is added in version 3.0 in order to plot a confusion matrix using Matplotlib or Seaborn.

>>> cm.plot()

>>> from matplotlib import pyplot as plt
>>> cm.plot(cmap=plt.cm.Greens,number_label=True,plot_lib="matplotlib")

>>> cm.plot(cmap=plt.cm.Reds,normalized=True,number_label=True,plot_lib="seaborn")

Online help

online_help function is added in version 1.1 in order to open each statistics definition in web browser

>>> from pycm import online_help
>>> online_help("J")
>>> online_help("SOA1(Landis & Koch)")
>>> online_help(2)

• List of items are available by calling online_help() (without argument)
• If PyCM website is not available, set alt_link = True (new in version 2.4)

Parameter recommender

This option has been added in version 1.9 to recommend the most related parameters considering the characteristics of the input dataset. The suggested parameters are selected according to some characteristics of the input such as being balance/imbalance and binary/multi-class. All suggestions can be categorized into three main groups: imbalanced dataset, binary classification for a balanced dataset, and multi-class classification for a balanced dataset. The recommendation lists have been gathered according to the respective paper of each parameter and the capabilities which had been claimed by the paper.

>>> cm.imbalance
False
>>> cm.binary
False
>>> cm.recommended_list
['MCC', 'TPR Micro', 'ACC', 'PPV Macro', 'BCD', 'Overall MCC', 'Hamming Loss', 'TPR Macro', 'Zero-one Loss', 'ERR', 'PPV Micro', 'Overall ACC']

Compare

In version 2.0, a method for comparing several confusion matrices is introduced. This option is a combination of several overall and class-based benchmarks. Each of the benchmarks evaluates the performance of the classification algorithm from good to poor and give them a numeric score. The score of good and poor performances are 1 and 0, respectively.

After that, two scores are calculated for each confusion matrices, overall and class-based. The overall score is the average of the score of six overall benchmarks which are Landis & Koch, Fleiss, Altman, Cicchetti, Cramer, and Matthews. In the same manner, the class-based score is the average of the score of six class-based benchmarks which are Positive Likelihood Ratio Interpretation, Negative Likelihood Ratio Interpretation, Discriminant Power Interpretation, AUC value Interpretation, Matthews Correlation Coefficient Interpretation and Yule's Q Interpretation. It should be noticed that if one of the benchmarks returns none for one of the classes, that benchmarks will be eliminated in total averaging. If the user sets weights for the classes, the averaging over the value of class-based benchmark scores will transform to a weighted average.

If the user sets the value of by_class boolean input True, the best confusion matrix is the one with the maximum class-based score. Otherwise, if a confusion matrix obtains the maximum of both overall and class-based scores, that will be reported as the best confusion matrix, but in any other case, the compared object doesn’t select the best confusion matrix.

>>> cm2 = ConfusionMatrix(matrix={0:{0:2,1:50,2:6},1:{0:5,1:50,2:3},2:{0:1,1:7,2:50}})
>>> cm3 = ConfusionMatrix(matrix={0:{0:50,1:2,2:6},1:{0:50,1:5,2:3},2:{0:1,1:55,2:2}})
>>> cp = Compare({"cm2":cm2,"cm3":cm3})
>>> print(cp)
Best : cm2
Rank Name Class-Score Overall-Score
1 cm2 9.05 2.55
2 cm3 6.05 1.98333
>>> cp.best
pycm.ConfusionMatrix(classes: [0, 1, 2])
>>> cp.sorted
['cm2', 'cm3']
>>> cp.best_name
'cm2'

Acceptable data types

ConfusionMatrix

1. actual_vector : python list or numpy array of any stringable objects
2. predict_vector : python list or numpy array of any stringable objects
3. matrix : dict
4. digit: int
5. threshold : FunctionType (function or lambda)
6. file : File object
7. sample_weight : python list or numpy array of numbers
8. transpose : bool
9. classes : python list

• Run help(ConfusionMatrix) for ConfusionMatrix object details

Compare

1. cm_dict : python dict of ConfusionMatrix object (str : ConfusionMatrix)
2. by_class : bool
3. weight : python dict of class weights (class_name : float)
4. digit: int

• Run help(Compare) for Compare object details

For more information visit here

Try PyCM in your browser!

PyCM can be used online in interactive Jupyter Notebooks via the Binder service! Try it out now! :

• Check Examples in Document folder

Issues & bug reports

Just fill an issue and describe it. We'll check it ASAP! or send an email to info@pycm.ir.

• Please complete the issue template

Outputs

1. HTML
2. CSV
3. PyCM
4. OBJ
5. COMP