{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "

Please cite us if you use the software

" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Example-5 (Sample weights)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Environment check" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Checking that the notebook is running on Google Colab or not." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import sys\n", "try:\n", " import google.colab\n", " !{sys.executable} -m pip -q -q install pycm\n", "except:\n", " pass" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Without weights" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "from pycm import ConfusionMatrix\n", "y_test = [2, 0, 2, 2, 0, 1, 1, 2, 2, 0, 1, 2]\n", "y_pred = [0, 0, 2, 1, 0, 2, 1, 0, 2, 0, 2, 2]" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "pycm.ConfusionMatrix(classes: [0, 1, 2])" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cm1=ConfusionMatrix(y_test, y_pred)\n", "cm1" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Predict 0 1 2 \n", "Actual\n", "0 3 0 0 \n", "\n", "1 0 1 2 \n", "\n", "2 2 1 3 \n", "\n", "\n", "\n", "\n", "\n", "Overall Statistics : \n", "\n", "95% CI (0.30439,0.86228)\n", "ACC Macro 0.72222\n", "ARI 0.09206\n", "AUNP 0.66667\n", "AUNU 0.69444\n", "Bangdiwala B 0.37255\n", "Bennett S 0.375\n", "CBA 0.47778\n", "CSI 0.17778\n", "Chi-Squared 6.6\n", "Chi-Squared DF 4\n", "Conditional Entropy 0.95915\n", "Cramer V 0.5244\n", "Cross Entropy 1.59352\n", "F1 Macro 0.56515\n", "F1 Micro 0.58333\n", "FNR Macro 0.38889\n", "FNR Micro 0.41667\n", "FPR Macro 0.22222\n", "FPR Micro 0.20833\n", "Gwet AC1 0.38931\n", "Hamming Loss 0.41667\n", "Joint Entropy 2.45915\n", "KL Divergence 0.09352\n", "Kappa 0.35484\n", "Kappa 95% CI (-0.07708,0.78675)\n", "Kappa No Prevalence 0.16667\n", "Kappa Standard Error 0.22036\n", "Kappa Unbiased 0.34426\n", "Krippendorff Alpha 0.37158\n", "Lambda A 0.16667\n", "Lambda B 0.42857\n", "Mutual Information 0.52421\n", "NIR 0.5\n", "NPV Macro 0.79048\n", "NPV Micro 0.79167\n", "Overall ACC 0.58333\n", "Overall CEN 0.46381\n", "Overall J (1.225,0.40833)\n", "Overall MCC 0.36667\n", "Overall MCEN 0.51894\n", "Overall RACC 0.35417\n", "Overall RACCU 0.36458\n", "P-Value 0.38721\n", "PPV Macro 0.56667\n", "PPV Micro 0.58333\n", "Pearson C 0.59568\n", "Phi-Squared 0.55\n", "RCI 0.34947\n", "RR 4.0\n", "Reference Entropy 1.5\n", "Response Entropy 1.48336\n", "SOA1(Landis & Koch) Fair\n", "SOA2(Fleiss) Poor\n", "SOA3(Altman) Fair\n", "SOA4(Cicchetti) Poor\n", "SOA5(Cramer) Relatively Strong\n", "SOA6(Matthews) Weak\n", "SOA7(Lambda A) Very Weak\n", "SOA8(Lambda B) Moderate\n", "SOA9(Krippendorff Alpha) Low\n", "SOA10(Pearson C) Strong\n", "Scott PI 0.34426\n", "Standard Error 0.14232\n", "TNR Macro 0.77778\n", "TNR Micro 0.79167\n", "TPR Macro 0.61111\n", "TPR Micro 0.58333\n", "Zero-one Loss 5\n", "\n", "Class Statistics :\n", "\n", "Classes 0 1 2 \n", "ACC(Accuracy) 0.83333 0.75 0.58333 \n", "AGF(Adjusted F-score) 0.9136 0.53995 0.5516 \n", "AGM(Adjusted geometric mean) 0.83729 0.692 0.60712 \n", "AM(Difference between automatic and manual classification) 2 -1 -1 \n", "AUC(Area under the ROC curve) 0.88889 0.61111 0.58333 \n", "AUCI(AUC value interpretation) Very Good Fair Poor \n", "AUPR(Area under the PR curve) 0.8 0.41667 0.55 \n", "BB(Braun-Blanquet similarity) 0.6 0.33333 0.5 \n", "BCD(Bray-Curtis dissimilarity) 0.08333 0.04167 0.04167 \n", "BM(Informedness or bookmaker informedness) 0.77778 0.22222 0.16667 \n", "CEN(Confusion entropy) 0.25 0.49658 0.60442 \n", "DOR(Diagnostic odds ratio) None 4.0 2.0 \n", "DP(Discriminant power) None 0.33193 0.16597 \n", "DPI(Discriminant power interpretation) None Poor Poor \n", "ERR(Error rate) 0.16667 0.25 0.41667 \n", "F0.5(F0.5 score) 0.65217 0.45455 0.57692 \n", "F1(F1 score - harmonic mean of precision and sensitivity) 0.75 0.4 0.54545 \n", "F2(F2 score) 0.88235 0.35714 0.51724 \n", "FDR(False discovery rate) 0.4 0.5 0.4 \n", "FN(False negative/miss/type 2 error) 0 2 3 \n", "FNR(Miss rate or false negative rate) 0.0 0.66667 0.5 \n", "FOR(False omission rate) 0.0 0.2 0.42857 \n", "FP(False positive/type 1 error/false alarm) 2 1 2 \n", "FPR(Fall-out or false positive rate) 0.22222 0.11111 0.33333 \n", "G(G-measure geometric mean of precision and sensitivity) 0.7746 0.40825 0.54772 \n", "GI(Gini index) 0.77778 0.22222 0.16667 \n", "GM(G-mean geometric mean of specificity and sensitivity) 0.88192 0.54433 0.57735 \n", "HD(Hamming distance) 2 3 5 \n", "IBA(Index of balanced accuracy) 0.95062 0.13169 0.27778 \n", "ICSI(Individual classification success index) 0.6 -0.16667 0.1 \n", "IS(Information score) 1.26303 1.0 0.26303 \n", "J(Jaccard index) 0.6 0.25 0.375 \n", "LS(Lift score) 2.4 2.0 1.2 \n", "MCC(Matthews correlation coefficient) 0.68313 0.2582 0.16903 \n", "MCCI(Matthews correlation coefficient interpretation) Moderate Negligible Negligible \n", "MCEN(Modified confusion entropy) 0.26439 0.5 0.6875 \n", "MK(Markedness) 0.6 0.3 0.17143 \n", "N(Condition negative) 9 9 6 \n", "NLR(Negative likelihood ratio) 0.0 0.75 0.75 \n", "NLRI(Negative likelihood ratio interpretation) Good Negligible Negligible \n", "NPV(Negative predictive value) 1.0 0.8 0.57143 \n", "OC(Overlap coefficient) 1.0 0.5 0.6 \n", "OOC(Otsuka-Ochiai coefficient) 0.7746 0.40825 0.54772 \n", "OP(Optimized precision) 0.70833 0.29545 0.44048 \n", "P(Condition positive or support) 3 3 6 \n", "PLR(Positive likelihood ratio) 4.5 3.0 1.5 \n", "PLRI(Positive likelihood ratio interpretation) Poor Poor Poor \n", "POP(Population) 12 12 12 \n", "PPV(Precision or positive predictive value) 0.6 0.5 0.6 \n", "PRE(Prevalence) 0.25 0.25 0.5 \n", "Q(Yule Q - coefficient of colligation) None 0.6 0.33333 \n", "QI(Yule Q interpretation) None Moderate Weak \n", "RACC(Random accuracy) 0.10417 0.04167 0.20833 \n", "RACCU(Random accuracy unbiased) 0.11111 0.0434 0.21007 \n", "TN(True negative/correct rejection) 7 8 4 \n", "TNR(Specificity or true negative rate) 0.77778 0.88889 0.66667 \n", "TON(Test outcome negative) 7 10 7 \n", "TOP(Test outcome positive) 5 2 5 \n", "TP(True positive/hit) 3 1 3 \n", "TPR(Sensitivity, recall, hit rate, or true positive rate) 1.0 0.33333 0.5 \n", "Y(Youden index) 0.77778 0.22222 0.16667 \n", "dInd(Distance index) 0.22222 0.67586 0.60093 \n", "sInd(Similarity index) 0.84287 0.52209 0.57508 \n", "\n" ] } ], "source": [ "print(cm1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## With random weights" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "from random import randint,seed\n", "seed(100)\n", "weights = [randint(1, 10) for i in range(len(y_test))]\n", "weights[2]*=9" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "pycm.ConfusionMatrix(classes: [0, 1, 2])" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cm2=ConfusionMatrix(y_test, y_pred, sample_weight=weights)\n", "cm2" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Predict 0 1 2 \n", "Actual\n", "0 24 0 0 \n", "\n", "1 0 7 8 \n", "\n", "2 12 3 76 \n", "\n", "\n", "\n", "\n", "\n", "Overall Statistics : \n", "\n", "95% CI (0.75748,0.88868)\n", "ACC Macro 0.88205\n", "ARI 0.48768\n", "AUNP 0.82779\n", "AUNU 0.82623\n", "Bangdiwala B 0.73932\n", "Bennett S 0.73462\n", "CBA 0.65617\n", "CSI 0.52442\n", "Chi-Squared 110.9678\n", "Chi-Squared DF 4\n", "Conditional Entropy 0.65034\n", "Cramer V 0.6533\n", "Cross Entropy 1.21\n", "F1 Macro 0.74286\n", "F1 Micro 0.82308\n", "FNR Macro 0.23272\n", "FNR Micro 0.17692\n", "FPR Macro 0.11481\n", "FPR Micro 0.08846\n", "Gwet AC1 0.76652\n", "Hamming Loss 0.17692\n", "Joint Entropy 1.82\n", "KL Divergence 0.04034\n", "Kappa 0.63722\n", "Kappa 95% CI (0.50272,0.77173)\n", "Kappa No Prevalence 0.64615\n", "Kappa Standard Error 0.06863\n", "Kappa Unbiased 0.63481\n", "Krippendorff Alpha 0.63621\n", "Lambda A 0.41026\n", "Lambda B 0.52174\n", "Mutual Information 0.5544\n", "NIR 0.7\n", "NPV Macro 0.86908\n", "NPV Micro 0.91154\n", "Overall ACC 0.82308\n", "Overall CEN 0.28807\n", "Overall J (1.82323,0.60774)\n", "Overall MCC 0.64625\n", "Overall MCEN 0.38181\n", "Overall RACC 0.51231\n", "Overall RACCU 0.51553\n", "P-Value 0.00097\n", "PPV Macro 0.75714\n", "PPV Micro 0.82308\n", "Pearson C 0.67861\n", "Phi-Squared 0.8536\n", "RCI 0.47399\n", "RR 43.33333\n", "Reference Entropy 1.16966\n", "Response Entropy 1.20474\n", "SOA1(Landis & Koch) Substantial\n", "SOA2(Fleiss) Intermediate to Good\n", "SOA3(Altman) Good\n", "SOA4(Cicchetti) Good\n", "SOA5(Cramer) Strong\n", "SOA6(Matthews) Moderate\n", "SOA7(Lambda A) Moderate\n", "SOA8(Lambda B) Moderate\n", "SOA9(Krippendorff Alpha) Low\n", "SOA10(Pearson C) Strong\n", "Scott PI 0.63481\n", "Standard Error 0.03347\n", "TNR Macro 0.88519\n", "TNR Micro 0.91154\n", "TPR Macro 0.76728\n", "TPR Micro 0.82308\n", "Zero-one Loss 23\n", "\n", "Class Statistics :\n", "\n", "Classes 0 1 2 \n", "ACC(Accuracy) 0.90769 0.91538 0.82308 \n", "AGF(Adjusted F-score) 0.94152 0.68599 0.76783 \n", "AGM(Adjusted geometric mean) 0.91704 0.81486 0.81018 \n", "AM(Difference between automatic and manual classification) 12 -5 -7 \n", "AUC(Area under the ROC curve) 0.9434 0.72029 0.81502 \n", "AUCI(AUC value interpretation) Excellent Good Very Good \n", "AUPR(Area under the PR curve) 0.83333 0.58333 0.86996 \n", "BB(Braun-Blanquet similarity) 0.66667 0.46667 0.83516 \n", "BCD(Bray-Curtis dissimilarity) 0.04615 0.01923 0.02692 \n", "BM(Informedness or bookmaker informedness) 0.88679 0.44058 0.63004 \n", "CEN(Confusion entropy) 0.23219 0.44655 0.28458 \n", "DOR(Diagnostic odds ratio) None 32.66667 19.63333 \n", "DP(Discriminant power) None 0.83477 0.71286 \n", "DPI(Discriminant power interpretation) None Poor Poor \n", "ERR(Error rate) 0.09231 0.08462 0.17692 \n", "F0.5(F0.5 score) 0.71429 0.63636 0.88993 \n", "F1(F1 score - harmonic mean of precision and sensitivity) 0.8 0.56 0.86857 \n", "F2(F2 score) 0.90909 0.5 0.84821 \n", "FDR(False discovery rate) 0.33333 0.3 0.09524 \n", "FN(False negative/miss/type 2 error) 0 8 15 \n", "FNR(Miss rate or false negative rate) 0.0 0.53333 0.16484 \n", "FOR(False omission rate) 0.0 0.06667 0.32609 \n", "FP(False positive/type 1 error/false alarm) 12 3 8 \n", "FPR(Fall-out or false positive rate) 0.11321 0.02609 0.20513 \n", "G(G-measure geometric mean of precision and sensitivity) 0.8165 0.57155 0.86927 \n", "GI(Gini index) 0.88679 0.44058 0.63004 \n", "GM(G-mean geometric mean of specificity and sensitivity) 0.9417 0.67416 0.81477 \n", "HD(Hamming distance) 12 11 23 \n", "IBA(Index of balanced accuracy) 0.98718 0.22395 0.6906 \n", "ICSI(Individual classification success index) 0.66667 0.16667 0.73993 \n", "IS(Information score) 1.85244 2.6009 0.37018 \n", "J(Jaccard index) 0.66667 0.38889 0.76768 \n", "LS(Lift score) 3.61111 6.06667 1.29252 \n", "MCC(Matthews correlation coefficient) 0.76889 0.52824 0.60381 \n", "MCCI(Matthews correlation coefficient interpretation) Strong Moderate Moderate \n", "MCEN(Modified confusion entropy) 0.26416 0.4754 0.40758 \n", "MK(Markedness) 0.66667 0.63333 0.57867 \n", "N(Condition negative) 106 115 39 \n", "NLR(Negative likelihood ratio) 0.0 0.54762 0.20737 \n", "NLRI(Negative likelihood ratio interpretation) Good Negligible Poor \n", "NPV(Negative predictive value) 1.0 0.93333 0.67391 \n", "OC(Overlap coefficient) 1.0 0.7 0.90476 \n", "OOC(Otsuka-Ochiai coefficient) 0.8165 0.57155 0.86927 \n", "OP(Optimized precision) 0.84769 0.56327 0.79836 \n", "P(Condition positive or support) 24 15 91 \n", "PLR(Positive likelihood ratio) 8.83333 17.88889 4.07143 \n", "PLRI(Positive likelihood ratio interpretation) Fair Good Poor \n", "POP(Population) 130 130 130 \n", "PPV(Precision or positive predictive value) 0.66667 0.7 0.90476 \n", "PRE(Prevalence) 0.18462 0.11538 0.7 \n", "Q(Yule Q - coefficient of colligation) None 0.94059 0.90307 \n", "QI(Yule Q interpretation) None Strong Strong \n", "RACC(Random accuracy) 0.05112 0.00888 0.45231 \n", "RACCU(Random accuracy unbiased) 0.05325 0.00925 0.45303 \n", "TN(True negative/correct rejection) 94 112 31 \n", "TNR(Specificity or true negative rate) 0.88679 0.97391 0.79487 \n", "TON(Test outcome negative) 94 120 46 \n", "TOP(Test outcome positive) 36 10 84 \n", "TP(True positive/hit) 24 7 76 \n", "TPR(Sensitivity, recall, hit rate, or true positive rate) 1.0 0.46667 0.83516 \n", "Y(Youden index) 0.88679 0.44058 0.63004 \n", "dInd(Distance index) 0.11321 0.53397 0.26315 \n", "sInd(Similarity index) 0.91995 0.62243 0.81392 \n", "\n" ] } ], "source": [ "print(cm2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n", "\n", "" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.5.2" } }, "nbformat": 4, "nbformat_minor": 2 }