diff --git a/Maternal health Risk Prediction/Maternal Health Risk Data Set.csv b/Maternal health Risk Prediction/Maternal Health Risk Data Set.csv new file mode 100644 index 00000000..d58d799f --- /dev/null +++ b/Maternal health Risk Prediction/Maternal Health Risk Data Set.csv @@ -0,0 +1,1015 @@ +Age,SystolicBP,DiastolicBP,BS,BodyTemp,HeartRate,RiskLevel +25,130,80,15,98,86,high risk +35,140,90,13,98,70,high risk +29,90,70,8,100,80,high risk +30,140,85,7,98,70,high risk +35,120,60,6.1,98,76,low risk +23,140,80,7.01,98,70,high risk +23,130,70,7.01,98,78,mid risk +35,85,60,11,102,86,high risk +32,120,90,6.9,98,70,mid risk +42,130,80,18,98,70,high risk +23,90,60,7.01,98,76,low risk +19,120,80,7,98,70,mid risk +25,110,89,7.01,98,77,low risk +20,120,75,7.01,100,70,mid risk +48,120,80,11,98,88,mid risk +15,120,80,7.01,98,70,low risk +50,140,90,15,98,90,high risk +25,140,100,7.01,98,80,high risk +30,120,80,6.9,101,76,mid risk +10,70,50,6.9,98,70,low risk +40,140,100,18,98,90,high risk +50,140,80,6.7,98,70,mid risk +21,90,65,7.5,98,76,low risk +18,90,60,7.5,98,70,low risk +21,120,80,7.5,98,76,low risk +16,100,70,7.2,98,80,low risk +19,120,75,7.2,98,66,low risk +22,100,65,7.2,98,70,low risk +49,120,90,7.2,98,77,low risk +28,90,60,7.2,98,82,low risk +20,100,90,7.1,98,88,low risk +23,100,85,7.1,98,66,low risk +22,120,90,7.1,98,82,low risk +21,120,80,7.1,98,77,low risk +21,75,50,6.1,98,70,low risk +12,95,60,6.1,102,60,low risk +60,120,80,6.1,98,75,low risk +55,100,65,6.1,98,66,low risk +45,120,95,6.1,98,66,low risk +35,100,70,6.1,98,66,low risk +22,120,85,6.1,98,88,low risk +23,120,90,6.1,98,60,low risk +25,90,70,6.1,98,80,low risk +30,120,80,6.1,98,70,low risk +23,120,90,6.1,98,70,low risk +32,120,90,7.5,98,70,low risk +42,120,80,7.5,98,70,low risk +23,90,60,7.5,98,76,low risk +15,76,49,7.5,98,77,low risk +15,120,80,7,98,70,low risk +25,120,80,7,98,66,low risk +22,100,65,7,98,80,low risk +35,100,70,7,98,60,low risk +19,120,85,7,98,60,low risk +60,90,65,7,98,77,low risk +23,120,90,6.7,98,70,low risk +32,120,90,6.4,98,70,low risk +42,120,80,6.4,98,70,low risk +23,90,60,6.4,98,76,low risk +15,76,49,6.4,98,77,low risk +15,120,80,7.2,98,70,low risk +15,80,60,7,98,80,low risk +12,95,60,7.2,98,77,low risk +29,90,70,6.7,98,80,mid risk +31,120,60,6.1,98,76,mid risk +29,130,70,6.7,98,78,mid risk +17,85,60,9,102,86,mid risk +19,120,80,7,98,70,mid risk +20,110,60,7,100,70,mid risk +32,120,65,6,101,76,mid risk +26,85,60,6,101,86,mid risk +29,130,70,7.7,98,78,mid risk +19,120,80,7,98,70,mid risk +54,130,70,12,98,67,mid risk +44,120,90,16,98,80,mid risk +23,130,70,6.9,98,70,mid risk +22,85,60,6.9,98,76,mid risk +55,120,90,12,98,70,mid risk +35,120,80,6.9,98,78,mid risk +21,90,60,6.9,98,86,mid risk +16,90,65,6.9,98,76,mid risk +33,115,65,7,98,70,mid risk +12,95,60,6.9,98,65,mid risk +28,120,90,6.9,98,70,mid risk +21,90,65,6.9,98,76,mid risk +18,90,60,6.9,98,70,mid risk +21,120,80,6.9,98,76,mid risk +16,100,70,6.9,98,80,mid risk +19,120,75,6.9,98,66,mid risk +23,100,85,6.9,98,66,mid risk +22,120,90,7.8,98,82,mid risk +60,120,85,15,98,60,mid risk +13,90,65,7.8,101,80,mid risk +23,120,90,7.8,98,60,mid risk +28,115,60,7.8,101,86,mid risk +50,120,80,7.8,98,70,mid risk +29,130,70,7.8,98,78,mid risk +19,120,80,7,98,70,mid risk +19,120,85,7.8,98,60,mid risk +60,90,65,6.8,98,77,mid risk +55,120,90,6.8,98,66,mid risk +25,120,80,6.8,98,66,mid risk +48,140,90,15,98,90,high risk +25,140,100,6.8,98,80,high risk +23,140,90,6.8,98,70,high risk +34,85,60,11,102,86,high risk +50,140,90,15,98,90,high risk +25,140,100,6.8,98,80,high risk +42,140,100,18,98,90,high risk +32,140,100,7.9,98,78,high risk +50,140,95,17,98,60,high risk +38,135,60,7.9,101,86,high risk +39,90,70,9,98,80,high risk +30,140,100,15,98,70,high risk +63,140,90,15,98,90,high risk +25,140,100,7.9,98,80,high risk +30,120,80,7.9,101,76,high risk +55,140,100,18,98,90,high risk +32,140,100,7.9,98,78,high risk +30,140,100,15,98,70,high risk +48,120,80,11,98,88,high risk +49,140,90,15,98,90,high risk +25,140,100,7.5,98,80,high risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +35,140,100,7.5,98,66,high risk +54,140,100,15,98,66,high risk +55,140,95,19,98,77,high risk +29,120,70,9,98,80,high risk +48,120,80,11,98,88,high risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +35,140,100,7.5,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,high risk +22,90,60,7.5,102,60,high risk +40,120,85,15,98,60,high risk +55,140,95,19,98,77,high risk +50,130,100,16,98,75,high risk +18,120,80,6.9,102,76,mid risk +32,140,100,6.9,98,78,high risk +17,90,60,6.9,101,76,mid risk +17,90,63,6.9,101,70,mid risk +25,120,90,6.7,101,80,mid risk +17,120,80,6.7,102,76,mid risk +14,90,65,7,101,70,high risk +15,80,60,6.7,98,80,low risk +15,100,65,6.7,98,76,low risk +12,95,60,6.7,98,77,low risk +37,120,90,11,98,88,high risk +18,100,70,6.7,98,76,low risk +21,100,85,6.7,98,70,low risk +17,110,75,12,101,76,high risk +25,120,90,7.5,98,80,low risk +23,85,65,7.5,98,70,low risk +12,95,60,7.5,98,65,low risk +28,120,90,7.5,98,70,low risk +40,120,90,12,98,80,high risk +55,129,85,7.5,98,88,low risk +25,100,90,7.5,98,76,low risk +35,120,80,7.5,98,80,low risk +21,90,65,7.5,98,76,low risk +18,90,60,7.5,98,70,low risk +21,120,80,7.5,98,76,low risk +16,100,70,7.2,98,80,low risk +19,120,75,7.2,98,66,low risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +22,100,65,7.2,98,70,low risk +49,120,90,7.2,98,77,low risk +28,90,60,7.2,98,82,low risk +12,90,60,7.9,102,66,high risk +20,100,90,7.1,98,88,low risk +23,100,85,7.1,98,66,low risk +22,120,90,7.1,98,82,low risk +21,120,80,7.1,98,77,low risk +35,140,100,8,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,high risk +21,75,50,6.1,98,70,low risk +12,95,60,6.1,102,60,low risk +60,120,85,15,98,60,high risk +55,140,95,19,98,77,high risk +50,130,100,16,98,75,high risk +60,120,80,6.1,98,75,low risk +55,100,65,6.1,98,66,low risk +45,120,95,6.1,98,66,low risk +35,100,70,6.1,98,66,low risk +22,120,85,6.1,98,88,low risk +13,90,65,7.9,101,80,mid risk +23,120,90,6.1,98,60,low risk +17,90,65,6.1,103,67,high risk +28,83,60,8,101,86,high risk +50,120,80,15,98,70,high risk +25,90,70,6.1,98,80,low risk +30,120,80,6.1,98,70,low risk +31,120,60,6.1,98,76,mid risk +23,120,90,6.1,98,70,low risk +29,130,70,6.1,98,78,mid risk +17,85,60,9,102,86,high risk +32,120,90,7.5,98,70,low risk +42,120,80,7.5,98,70,low risk +23,90,60,7.5,98,76,low risk +19,120,80,7,98,70,mid risk +15,76,49,7.5,98,77,low risk +33,120,75,10,98,70,high risk +48,120,80,11,98,88,high risk +15,120,80,7,98,70,low risk +25,120,80,7,98,66,low risk +22,100,65,7,98,80,low risk +50,140,95,17,98,60,high risk +35,100,70,7,98,60,low risk +19,120,85,7,98,60,low risk +60,90,65,7,98,77,low risk +28,85,60,9,101,86,mid risk +50,140,80,6.7,98,70,mid risk +29,90,70,6.7,98,80,mid risk +30,140,100,15,98,70,high risk +31,120,60,6.1,98,76,mid risk +23,120,90,6.7,98,70,low risk +29,130,70,6.7,98,78,mid risk +17,85,60,9,102,86,mid risk +32,120,90,6.4,98,70,low risk +42,120,80,6.4,98,70,low risk +23,90,60,6.4,98,76,low risk +19,120,80,7,98,70,mid risk +15,76,49,6.4,98,77,low risk +29,120,75,7.2,100,70,high risk +48,120,80,11,98,88,high risk +15,120,80,7.2,98,70,low risk +50,140,90,15,98,77,high risk +25,140,100,7.2,98,80,high risk +55,140,80,7.2,101,76,high risk +20,110,60,7,100,70,mid risk +40,140,100,18,98,77,high risk +28,120,80,9,102,76,high risk +32,140,100,8,98,70,high risk +17,90,60,11,101,78,high risk +17,90,63,8,101,70,high risk +25,120,90,12,101,80,high risk +17,120,80,7,102,76,high risk +19,90,65,11,101,70,high risk +15,80,60,7,98,80,low risk +32,120,65,6,101,76,mid risk +12,95,60,7.2,98,77,low risk +37,120,90,11,98,88,high risk +18,100,70,6.8,98,76,low risk +21,100,85,6.9,98,70,low risk +17,110,75,13,101,76,high risk +25,120,90,15,98,80,high risk +10,85,65,6.9,98,70,low risk +12,95,60,6.9,98,65,low risk +28,120,90,6.9,98,70,low risk +40,120,90,6.9,98,80,low risk +55,110,85,6.9,98,88,low risk +25,100,90,6.9,98,76,low risk +35,120,80,6.9,98,80,low risk +21,90,65,6.9,98,76,low risk +18,90,60,6.9,98,70,low risk +21,120,80,6.9,98,76,low risk +16,100,70,6.9,98,80,low risk +19,120,75,6.9,98,66,low risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +22,100,65,6.9,98,70,low risk +49,120,90,6.9,98,77,low risk +28,90,60,6.9,98,82,low risk +12,90,60,8,102,66,high risk +20,100,90,7,98,88,low risk +23,100,85,7,98,66,low risk +22,120,90,7,98,82,low risk +21,120,80,7,98,77,low risk +35,140,100,9,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,high risk +21,75,50,7.7,98,60,low risk +12,90,60,11,102,60,high risk +60,120,85,15,98,60,high risk +55,140,95,19,98,77,high risk +50,130,100,16,98,76,high risk +60,120,80,7.7,98,75,low risk +55,100,65,7.7,98,66,low risk +45,120,95,7.7,98,66,low risk +35,100,70,7.7,98,66,low risk +22,120,85,7.7,98,88,low risk +13,90,65,9,101,80,high risk +23,120,90,7.7,98,60,low risk +17,90,65,7.7,103,67,high risk +26,85,60,6,101,86,mid risk +50,120,80,7.7,98,70,low risk +19,90,70,7.7,98,80,low risk +30,120,80,7.7,98,70,low risk +31,120,60,6.1,98,76,low risk +23,120,80,7.7,98,70,low risk +29,130,70,7.7,98,78,mid risk +17,85,60,6.3,102,86,high risk +32,120,90,7.7,98,70,low risk +42,120,80,7.7,98,70,low risk +23,90,60,7.7,98,76,low risk +19,120,80,7,98,70,mid risk +15,75,49,7.7,98,77,low risk +40,120,75,7.7,98,70,high risk +48,120,80,11,98,88,high risk +15,120,80,7.7,98,70,low risk +25,120,80,7.7,98,66,low risk +22,100,65,6.9,98,80,low risk +12,120,95,6.9,98,60,low risk +35,100,70,6.9,98,60,low risk +19,120,85,6.9,98,60,low risk +60,90,65,6.9,98,77,low risk +55,120,90,6.9,98,76,low risk +35,90,65,6.9,98,75,low risk +51,85,60,6.9,98,66,low risk +62,120,80,6.9,98,66,low risk +25,90,70,6.9,98,66,low risk +21,120,80,6.9,98,88,low risk +22,120,60,15,98,80,high risk +55,120,90,18,98,60,high risk +54,130,70,12,98,67,mid risk +35,85,60,19,98,86,high risk +43,120,90,18,98,70,high risk +12,120,80,6.9,98,80,low risk +65,90,60,6.9,98,70,low risk +60,120,80,6.9,98,76,low risk +25,120,90,6.9,98,70,low risk +22,90,65,6.9,98,78,low risk +66,85,60,6.9,98,86,low risk +56,120,80,13,98,70,high risk +35,90,70,6.9,98,70,low risk +43,120,80,15,98,76,high risk +35,120,60,6.9,98,70,low risk +44,120,90,16,98,80,mid risk +23,130,70,6.9,98,70,mid risk +22,85,60,6.9,98,76,mid risk +55,120,90,12,98,70,mid risk +35,120,80,6.9,98,78,mid risk +21,90,60,6.9,98,86,mid risk +45,120,80,6.9,103,70,low risk +70,85,60,6.9,102,70,low risk +65,120,90,6.9,103,76,low risk +55,120,80,6.9,102,80,low risk +45,90,60,18,101,70,high risk +22,120,80,6.9,103,76,low risk +16,90,65,6.9,98,76,mid risk +12,95,60,6.9,98,77,low risk +37,120,90,11,98,88,high risk +18,100,70,6.9,98,76,low risk +21,100,85,6.9,98,70,low risk +17,110,75,6.9,101,76,high risk +25,120,90,6.9,98,80,low risk +33,115,65,7,98,70,mid risk +12,95,60,6.9,98,65,mid risk +28,120,90,6.9,98,70,mid risk +40,120,90,6.9,98,80,high risk +55,110,85,6.9,98,88,high risk +25,100,90,6.9,98,76,high risk +35,120,80,6.9,98,80,high risk +21,90,65,6.9,98,76,mid risk +18,90,60,6.9,98,70,mid risk +21,120,80,6.9,98,76,mid risk +16,100,70,6.9,98,80,mid risk +19,120,75,6.9,98,66,mid risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +23,100,85,6.9,98,66,mid risk +22,120,90,7.8,98,82,mid risk +21,120,80,7.8,98,77,low risk +35,140,100,7.8,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,high risk +21,75,50,7.8,98,60,low risk +12,90,60,7.8,102,60,high risk +60,120,85,15,98,60,mid risk +55,140,95,19,98,77,high risk +50,130,100,16,98,75,high risk +60,120,80,7.8,98,75,high risk +55,100,65,7.8,98,66,low risk +45,120,95,7.8,98,66,low risk +35,100,70,7.8,98,66,low risk +22,120,85,7.8,98,88,low risk +13,90,65,7.8,101,80,mid risk +23,120,90,7.8,98,60,mid risk +17,90,65,7.8,103,67,high risk +28,115,60,7.8,101,86,mid risk +50,120,80,7.8,98,70,mid risk +19,90,70,7.8,98,80,low risk +30,120,80,7.8,98,70,low risk +31,120,60,6.1,98,76,low risk +23,120,70,7.8,98,70,low risk +29,130,70,7.8,98,78,mid risk +17,85,69,7.8,102,86,high risk +32,120,90,7.8,98,70,low risk +42,120,80,7.8,98,70,low risk +23,90,60,7.8,98,76,low risk +19,120,80,7,98,70,mid risk +15,76,49,7.8,98,77,low risk +20,120,75,7.8,98,70,low risk +48,120,80,11,98,88,high risk +15,120,80,7.8,98,70,low risk +25,120,80,7.8,98,66,low risk +22,100,65,7.8,98,80,low risk +12,120,95,7.8,98,60,low risk +35,100,70,7.8,98,60,low risk +19,120,85,7.8,98,60,mid risk +60,90,65,6.8,98,77,mid risk +55,120,90,6.8,98,66,mid risk +25,120,80,6.8,98,66,mid risk +22,100,65,6.8,98,88,low risk +12,120,95,6.8,98,60,mid risk +35,100,70,6.8,98,60,mid risk +19,120,90,6.8,98,60,mid risk +60,90,65,6.8,98,77,mid risk +55,120,90,6.8,98,78,low risk +50,130,80,16,102,76,mid risk +27,120,90,6.8,102,68,mid risk +60,140,90,12,98,77,high risk +55,100,70,6.8,101,80,mid risk +60,140,80,16,98,66,high risk +12,120,90,6.8,98,80,mid risk +17,140,100,6.8,103,80,high risk +60,120,80,6.8,98,77,mid risk +22,100,65,6.8,98,88,low risk +36,140,100,6.8,102,76,high risk +22,90,60,6.8,98,77,low risk +25,120,100,6.8,98,60,mid risk +35,100,60,15,98,80,high risk +40,140,100,13,101,66,high risk +27,120,70,6.8,98,77,low risk +36,140,100,6.8,102,76,high risk +22,90,60,6.8,98,77,mid risk +25,120,100,6.8,98,60,low risk +35,100,60,15,98,80,high risk +40,140,100,13,101,66,high risk +27,120,70,6.8,98,77,low risk +27,120,70,6.8,98,77,low risk +65,130,80,15,98,86,high risk +35,140,80,13,98,70,high risk +29,90,70,10,98,80,high risk +30,120,80,6.8,98,70,mid risk +35,120,60,6.1,98,76,mid risk +23,140,90,6.8,98,70,high risk +23,130,70,6.8,98,78,mid risk +35,85,60,11,102,86,high risk +32,120,90,6.8,98,70,low risk +43,130,80,18,98,70,mid risk +23,99,60,6.8,98,76,low risk +19,120,80,7,98,70,mid risk +15,76,49,6.8,98,77,low risk +30,120,75,6.8,98,70,mid risk +48,120,80,11,98,88,high risk +15,120,80,6.8,98,70,low risk +48,140,90,15,98,90,high risk +25,140,100,6.8,98,80,high risk +29,100,70,6.8,98,80,low risk +32,120,80,6.8,98,70,mid risk +35,120,60,6.1,98,76,low risk +23,140,90,6.8,98,70,high risk +23,130,70,6.8,98,78,mid risk +34,85,60,11,102,86,high risk +32,120,90,6.8,98,70,low risk +42,130,80,18,98,70,mid risk +23,90,60,6.8,98,76,low risk +19,120,80,7,98,70,mid risk +15,76,49,6.8,98,77,low risk +20,120,75,6.8,98,70,low risk +48,120,80,11,98,88,low risk +15,120,80,6.8,98,70,low risk +50,140,90,15,98,90,high risk +25,140,100,6.8,98,80,high risk +30,120,80,6.8,101,76,low risk +31,110,90,6.8,100,70,mid risk +42,140,100,18,98,90,high risk +18,120,80,6.8,102,76,low risk +32,140,100,7.9,98,78,high risk +17,90,60,7.9,101,76,low risk +19,120,80,7,98,70,mid risk +15,76,49,7.9,98,77,low risk +19,120,75,7.9,98,70,low risk +48,120,80,11,98,88,low risk +15,120,80,7.9,98,70,low risk +25,120,80,7.9,98,66,mid risk +22,100,65,7.9,98,80,low risk +50,140,95,17,98,60,high risk +35,100,70,7.9,98,60,low risk +19,120,85,7.9,98,60,low risk +60,90,65,7.9,98,77,low risk +38,135,60,7.9,101,86,high risk +50,120,80,7.9,98,70,low risk +39,90,70,9,98,80,high risk +30,140,100,15,98,70,high risk +31,120,60,6.1,98,76,mid risk +23,120,90,7.9,98,70,mid risk +29,130,70,7.9,98,78,mid risk +17,85,60,7.9,102,86,low risk +32,120,90,7.9,98,70,low risk +42,120,80,7.9,98,70,low risk +23,90,60,7.9,98,76,low risk +19,120,80,7,98,70,low risk +15,76,49,7.9,98,77,low risk +16,120,75,7.9,98,7,low risk +48,120,80,11,98,88,mid risk +15,120,80,7.9,98,70,low risk +63,140,90,15,98,90,high risk +25,140,100,7.9,98,80,high risk +30,120,80,7.9,101,76,high risk +17,70,50,7.9,98,70,low risk +55,140,100,18,98,90,high risk +18,120,80,7.9,102,76,mid risk +32,140,100,7.9,98,78,high risk +17,90,60,7.5,101,76,low risk +17,90,63,7.5,101,70,low risk +25,120,90,7.5,101,80,low risk +17,120,80,7.5,102,76,low risk +19,90,65,7.5,101,70,low risk +15,80,60,7.5,98,80,low risk +60,90,65,7.5,98,77,low risk +18,85,60,7.5,101,86,mid risk +50,120,80,7.5,98,70,low risk +19,90,70,7.5,98,80,low risk +30,140,100,15,98,70,high risk +31,120,60,6.1,98,76,low risk +23,120,90,7.5,98,70,low risk +29,130,70,7.5,98,78,mid risk +17,85,60,7.5,102,86,low risk +32,120,90,7.5,98,70,low risk +42,120,80,7.5,98,70,low risk +42,90,60,7.5,98,76,low risk +19,120,80,7,98,70,low risk +15,78,49,7.5,98,77,low risk +23,120,75,8,98,70,mid risk +48,120,80,11,98,88,high risk +15,120,80,7.5,98,70,mid risk +49,140,90,15,98,90,high risk +25,140,100,7.5,98,80,high risk +30,120,80,7.5,101,76,mid risk +16,70,50,7.5,100,70,low risk +16,100,70,7.5,98,80,low risk +19,120,75,7.5,98,66,low risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +22,100,65,7.5,98,70,low risk +49,120,90,7.5,98,77,low risk +28,90,60,7.5,98,82,low risk +12,90,60,7.5,102,66,low risk +20,100,90,7.5,98,88,low risk +23,100,85,7.5,98,66,low risk +22,120,90,7.5,98,82,low risk +21,120,80,7.5,98,77,low risk +35,140,100,7.5,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,mid risk +21,75,50,7.5,98,60,low risk +12,90,60,7.5,102,60,low risk +60,120,85,15,98,60,mid risk +55,140,95,19,98,77,high risk +50,130,100,16,98,75,mid risk +60,120,80,7.5,98,75,low risk +55,100,65,7.5,98,66,low risk +45,120,95,7.5,98,66,low risk +35,100,70,7.5,98,66,low risk +22,120,85,7.5,98,88,low risk +13,90,65,7.5,101,80,low risk +23,120,90,7.5,98,60,low risk +17,90,65,7.5,103,67,low risk +28,115,60,7.5,101,86,mid risk +59,120,80,7.5,98,70,low risk +29,120,70,9,98,80,high risk +23,120,80,7.5,98,70,low risk +31,120,60,6.1,98,76,mid risk +23,120,80,7.5,98,70,mid risk +29,130,70,7.5,98,78,mid risk +17,85,60,7.5,102,86,low risk +32,120,90,7.5,98,70,low risk +42,120,80,7.5,98,70,low risk +23,90,60,7.5,98,76,low risk +19,120,80,7,98,70,low risk +15,78,49,7.5,98,77,low risk +20,120,75,7.5,98,70,low risk +48,120,80,11,98,88,high risk +15,120,80,7.5,98,70,low risk +24,120,80,7.5,98,66,low risk +16,100,70,7.5,98,80,low risk +19,120,76,7.5,98,66,low risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +22,100,65,7.5,98,70,mid risk +49,120,90,7.5,98,77,mid risk +28,90,60,7.5,98,82,mid risk +12,90,60,7.5,102,66,mid risk +20,100,90,7.5,98,88,mid risk +23,100,85,7.5,98,66,mid risk +22,120,90,7.5,98,82,mid risk +21,120,80,7.5,98,77,mid risk +35,140,100,7.5,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,high risk +21,75,50,7.5,98,60,low risk +22,90,60,7.5,102,60,high risk +40,120,85,15,98,60,high risk +55,140,95,19,98,77,high risk +50,130,100,16,98,75,high risk +60,120,80,7.5,98,75,mid risk +40,120,85,15,98,60,high risk +55,140,95,19,98,77,high risk +50,130,100,16,98,75,mid risk +41,120,80,7.5,98,75,low risk +55,100,65,7.5,98,66,low risk +45,120,95,7.5,98,66,low risk +35,100,70,7.5,98,66,low risk +22,120,85,7.5,98,88,low risk +13,90,65,7.5,101,80,high risk +23,120,90,7.5,98,60,low risk +17,90,65,7.5,103,67,mid risk +27,135,60,7.5,101,86,high risk +50,120,80,15,98,70,high risk +34,110,70,7,98,80,high risk +32,120,80,7.5,98,70,low risk +31,120,60,6.1,98,76,low risk +23,120,90,7.5,98,70,low risk +29,130,70,7.5,98,78,mid risk +17,85,60,7.5,101,86,high risk +32,120,90,7.5,98,70,low risk +42,120,80,7.5,98,70,low risk +23,90,60,7.5,98,76,low risk +19,120,80,7,98,70,mid risk +15,76,49,7.5,98,77,low risk +20,120,76,7.5,98,70,low risk +48,120,80,11,98,88,high risk +15,120,80,7.5,98,70,low risk +24,120,80,7.5,98,66,low risk +22,100,65,12,98,80,high risk +50,140,95,17,98,60,high risk +35,100,70,11,98,60,high risk +19,120,85,9,98,60,mid risk +30,90,65,8,98,77,mid risk +28,85,60,9,101,86,mid risk +50,130,80,15,98,86,high risk +35,140,90,13,98,70,high risk +29,90,70,11,100,80,high risk +19,120,60,7,98.4,70,low risk +46,140,100,12,99,90,high risk +28,95,60,10,101,86,high risk +50,120,80,7,98,70,mid risk +39,110,70,7.9,98,80,mid risk +25,140,100,15,98.6,70,high risk +31,120,60,6.1,98,76,low risk +23,120,85,8,98,70,low risk +29,130,70,8,98,78,mid risk +17,90,60,9,102,86,mid risk +32,120,90,7,100,70,mid risk +42,120,90,9,98,70,mid risk +23,90,60,6.7,98,76,low risk +19,120,80,7,98,70,low risk +15,76,68,7,98,77,low risk +34,120,75,8,98,70,low risk +48,120,80,11,98,88,high risk +15,120,80,6.6,99,70,low risk +27,140,90,15,98,90,high risk +25,140,100,12,99,80,high risk +36,120,90,7,98,82,mid risk +30,120,80,9,101,76,mid risk +15,70,50,6,98,70,mid risk +40,120,95,7,98,70,high risk +15,90,60,6,98,80,low risk +21,90,50,6.9,98,60,low risk +15,90,49,6,98,77,low risk +21,90,50,6.5,98,60,low risk +15,90,49,6,98,77,low risk +15,90,49,6.7,99,77,low risk +15,90,49,6,99,77,low risk +10,100,50,6,99,70,mid risk +15,100,49,6.8,99,77,low risk +15,100,49,6,99,77,low risk +12,100,50,6.4,98,70,mid risk +15,100,60,6,98,80,low risk +35,140,90,13,98,70,high risk +29,90,70,8,100,80,high risk +30,140,85,7,98,70,high risk +23,140,80,7.01,98,70,high risk +35,85,60,11,102,86,high risk +42,130,80,18,98,70,high risk +50,140,90,15,98,90,high risk +25,140,100,7.01,98,80,high risk +40,140,100,18,98,90,high risk +32,140,100,6.9,98,78,high risk +14,90,65,7,101,70,high risk +37,120,90,11,98,88,high risk +17,110,75,12,101,76,high risk +40,120,90,12,98,80,high risk +40,160,100,19,98,77,high risk +20,120,76,7.5,98,70,low risk +15,120,80,7.5,98,70,low risk +24,120,80,7.5,98,66,low risk +19,120,60,7,98.4,70,low risk +31,120,60,6.1,98,76,low risk +23,120,85,8,98,70,low risk +23,90,60,6.7,98,76,low risk +19,120,80,7,98,70,low risk +15,76,68,7,98,77,low risk +34,120,75,8,98,70,low risk +15,120,80,6.6,99,70,low risk +15,90,60,6,98,80,low risk +21,90,50,6.9,98,60,low risk +15,100,49,7.6,98,77,low risk +12,100,50,6,98,70,mid risk +21,100,50,6.8,98,60,low risk +23,130,70,7.01,98,78,mid risk +32,120,90,6.9,98,70,mid risk +19,120,80,7,98,70,mid risk +20,120,75,7.01,100,70,mid risk +48,120,80,11,98,88,mid risk +30,120,80,6.9,101,76,mid risk +18,120,80,6.9,102,76,mid risk +17,90,60,6.9,101,76,mid risk +17,90,63,6.9,101,70,mid risk +25,120,90,6.7,101,80,mid risk +17,120,80,6.7,102,76,mid risk +13,90,65,7.9,101,80,mid risk +31,120,60,6.1,98,76,mid risk +29,130,70,6.1,98,78,mid risk +19,120,80,7,98,70,mid risk +28,85,60,9,101,86,mid risk +50,140,80,6.7,98,70,mid risk +29,90,70,6.7,98,80,mid risk +31,120,60,6.1,98,76,mid risk +29,130,70,6.7,98,78,mid risk +17,85,60,9,102,86,mid risk +19,120,80,7,98,70,mid risk +20,110,60,7,100,70,mid risk +19,120,80,7,98,70,mid risk +20,120,75,7.01,100,70,mid risk +48,120,80,11,98,88,mid risk +30,120,80,6.9,101,76,mid risk +18,120,80,6.9,102,76,mid risk +17,90,60,6.9,101,76,mid risk +17,90,63,6.9,101,70,mid risk +25,120,90,6.7,101,80,mid risk +17,120,80,6.7,102,76,mid risk +13,90,65,7.9,101,80,mid risk +31,120,60,6.1,98,76,mid risk +29,130,70,6.1,98,78,mid risk +19,120,80,7,98,70,mid risk +28,85,60,9,101,86,mid risk +50,140,80,6.7,98,70,mid risk +29,90,70,6.7,98,80,mid risk +31,120,60,6.1,98,76,mid risk +29,130,70,6.7,98,78,mid risk +17,85,60,9,102,86,mid risk +19,120,80,7,98,70,mid risk +20,110,60,7,100,70,mid risk +32,120,65,6,101,76,mid risk +26,85,60,6,101,86,mid risk +29,130,70,7.7,98,78,mid risk +19,120,80,7,98,70,mid risk +54,130,70,12,98,67,mid risk +44,120,90,16,98,80,mid risk +23,130,70,6.9,98,70,mid risk +22,85,60,6.9,98,76,mid risk +55,120,90,12,98,70,mid risk +35,120,80,6.9,98,78,mid risk +21,90,60,6.9,98,86,mid risk +16,90,65,6.9,98,76,mid risk +33,115,65,7,98,70,mid risk +12,95,60,6.9,98,65,mid risk +28,120,90,6.9,98,70,mid risk +21,90,65,6.9,98,76,mid risk +18,90,60,6.9,98,70,mid risk +21,120,80,6.9,98,76,mid risk +16,100,70,6.9,98,80,mid risk +19,120,75,6.9,98,66,mid risk +23,100,85,6.9,98,66,mid risk +22,120,90,7.8,98,82,mid risk +60,120,85,15,98,60,mid risk +13,90,65,7.8,101,80,mid risk +23,120,90,7.8,98,60,mid risk +28,115,60,7.8,101,86,mid risk +50,120,80,7.8,98,70,mid risk +29,130,70,7.8,98,78,mid risk +19,120,80,7,98,70,mid risk +19,120,85,7.8,98,60,mid risk +60,90,65,6.8,98,77,mid risk +55,120,90,6.8,98,66,mid risk +25,120,80,6.8,98,66,mid risk +12,120,95,6.8,98,60,mid risk +35,100,70,6.8,98,60,mid risk +19,120,90,6.8,98,60,mid risk +60,90,65,6.8,98,77,mid risk +50,130,80,16,102,76,mid risk +27,120,90,6.8,102,68,mid risk +55,100,70,6.8,101,80,mid risk +12,120,90,6.8,98,80,mid risk +60,120,80,6.8,98,77,mid risk +25,120,100,6.8,98,60,mid risk +22,90,60,6.8,98,77,mid risk +30,120,80,6.8,98,70,mid risk +35,120,60,6.1,98,76,mid risk +23,130,70,6.8,98,78,mid risk +43,130,80,18,98,70,mid risk +19,120,80,7,98,70,mid risk +30,120,75,6.8,98,70,mid risk +32,120,80,6.8,98,70,mid risk +23,130,70,6.8,98,78,mid risk +42,130,80,18,98,70,mid risk +19,120,80,7,98,70,mid risk +31,110,90,6.8,100,70,mid risk +19,120,80,7,98,70,mid risk +25,120,80,7.9,98,66,mid risk +31,120,60,6.1,98,76,mid risk +23,120,90,7.9,98,70,mid risk +29,130,70,7.9,98,78,mid risk +48,120,80,11,98,88,mid risk +18,120,80,7.9,102,76,mid risk +18,85,60,7.5,101,86,mid risk +29,130,70,7.5,98,78,mid risk +23,120,75,8,98,70,mid risk +15,120,80,7.5,98,70,mid risk +30,120,80,7.5,101,76,mid risk +40,120,95,11,98,80,mid risk +60,120,85,15,98,60,mid risk +50,130,100,16,98,75,mid risk +28,115,60,7.5,101,86,mid risk +31,120,60,6.1,98,76,mid risk +23,120,80,7.5,98,70,mid risk +29,130,70,7.5,98,78,mid risk +22,100,65,7.5,98,70,mid risk +49,120,90,7.5,98,77,mid risk +28,90,60,7.5,98,82,mid risk +12,90,60,7.5,102,66,mid risk +20,100,90,7.5,98,88,mid risk +23,100,85,7.5,98,66,mid risk +22,120,90,7.5,98,82,mid risk +21,120,80,7.5,98,77,mid risk +60,120,80,7.5,98,75,mid risk +50,130,100,16,98,75,mid risk +17,90,65,7.5,103,67,mid risk +29,130,70,7.5,98,78,mid risk +19,120,80,7,98,70,mid risk +19,120,85,9,98,60,mid risk +30,90,65,8,98,77,mid risk +28,85,60,9,101,86,mid risk +50,120,80,7,98,70,mid risk +39,110,70,7.9,98,80,mid risk +29,130,70,8,98,78,mid risk +17,90,60,9,102,86,mid risk +32,120,90,7,100,70,mid risk +42,120,90,9,98,70,mid risk +36,120,90,7,98,82,mid risk +30,120,80,9,101,76,mid risk +15,70,50,6,98,70,mid risk +10,100,50,6,99,70,mid risk +12,100,50,6.4,98,70,mid risk +12,100,50,6,98,70,mid risk +23,130,70,7.01,98,78,mid risk +32,120,90,6.9,98,70,mid risk +19,120,80,7,98,70,mid risk +20,120,75,7.01,100,70,mid risk +48,120,80,11,98,88,mid risk +30,120,80,6.9,101,76,mid risk +18,120,80,6.9,102,76,mid risk +17,90,60,6.9,101,76,mid risk +17,90,63,6.9,101,70,mid risk +25,120,90,6.7,101,80,mid risk +17,120,80,6.7,102,76,mid risk +13,90,65,7.9,101,80,mid risk +31,120,60,6.1,98,76,mid risk +29,130,70,6.1,98,78,mid risk +19,120,80,7,98,70,mid risk +28,85,60,9,101,86,mid risk +50,140,80,6.7,98,70,mid risk +29,90,70,6.7,98,80,mid risk +31,120,60,6.1,98,76,mid risk +29,130,70,6.7,98,78,mid risk +17,85,60,9,102,86,mid risk +19,120,80,7,98,70,mid risk +20,110,60,7,100,70,mid risk +32,120,65,6,101,76,mid risk +27,120,70,6.8,98,77,low risk +27,120,70,6.8,98,77,low risk +32,120,90,6.8,98,70,low risk +23,99,60,6.8,98,76,low risk +15,76,49,6.8,98,77,low risk +15,120,80,6.8,98,70,low risk +29,100,70,6.8,98,80,low risk +35,120,60,6.1,98,76,low risk +32,120,90,6.8,98,70,low risk +23,90,60,6.8,98,76,low risk +15,76,49,6.8,98,77,low risk +20,120,75,6.8,98,70,low risk +48,120,80,11,98,88,low risk +15,120,80,6.8,98,70,low risk +30,120,80,6.8,101,76,low risk +18,120,80,6.8,102,76,low risk +17,90,60,7.9,101,76,low risk +15,76,49,7.9,98,77,low risk +19,120,75,7.9,98,70,low risk +48,120,80,11,98,88,low risk +15,120,80,7.9,98,70,low risk +22,100,65,7.9,98,80,low risk +35,100,70,7.9,98,60,low risk +19,120,85,7.9,98,60,low risk +60,90,65,7.9,98,77,low risk +50,120,80,7.9,98,70,low risk +17,85,60,7.9,102,86,low risk +32,120,90,7.9,98,70,low risk +42,120,80,7.9,98,70,low risk +23,90,60,7.9,98,76,low risk +19,120,80,7,98,70,low risk +15,76,49,7.9,98,77,low risk +16,120,75,7.9,98,7,low risk +15,120,80,7.9,98,70,low risk +17,70,50,7.9,98,70,low risk +17,90,60,7.5,101,76,low risk +17,90,63,7.5,101,70,low risk +25,120,90,7.5,101,80,low risk +17,120,80,7.5,102,76,low risk +19,90,65,7.5,101,70,low risk +15,80,60,7.5,98,80,low risk +60,90,65,7.5,98,77,low risk +50,120,80,7.5,98,70,low risk +19,90,70,7.5,98,80,low risk +31,120,60,6.1,98,76,low risk +23,120,90,7.5,98,70,low risk +17,85,60,7.5,102,86,low risk +32,120,90,7.5,98,70,low risk +42,120,80,7.5,98,70,low risk +42,90,60,7.5,98,76,low risk +19,120,80,7,98,70,low risk +15,78,49,7.5,98,77,low risk +16,70,50,7.5,100,70,low risk +16,100,70,7.5,98,80,low risk +19,120,75,7.5,98,66,low risk +22,100,65,7.5,98,70,low risk +49,120,90,7.5,98,77,low risk +28,90,60,7.5,98,82,low risk +12,90,60,7.5,102,66,low risk +20,100,90,7.5,98,88,low risk +23,100,85,7.5,98,66,low risk +22,120,90,7.5,98,82,low risk +21,120,80,7.5,98,77,low risk +21,75,50,7.5,98,60,low risk +12,90,60,7.5,102,60,low risk +60,120,80,7.5,98,75,low risk +55,100,65,7.5,98,66,low risk +45,120,95,7.5,98,66,low risk +35,100,70,7.5,98,66,low risk +22,120,85,7.5,98,88,low risk +13,90,65,7.5,101,80,low risk +23,120,90,7.5,98,60,low risk +17,90,65,7.5,103,67,low risk +59,120,80,7.5,98,70,low risk +23,120,80,7.5,98,70,low risk +17,85,60,7.5,102,86,low risk +32,120,90,7.5,98,70,low risk +42,120,80,7.5,98,70,low risk +25,140,100,7.01,98,80,high risk +40,140,100,18,98,90,high risk +32,140,100,6.9,98,78,high risk +14,90,65,7,101,70,high risk +37,120,90,11,98,88,high risk +17,110,75,12,101,76,high risk +40,120,90,12,98,80,high risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +12,90,60,7.9,102,66,high risk +35,140,100,8,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,high risk +60,120,85,15,98,60,high risk +55,140,95,19,98,77,high risk +50,130,100,16,98,75,high risk +17,90,65,6.1,103,67,high risk +28,83,60,8,101,86,high risk +50,120,80,15,98,70,high risk +17,85,60,9,102,86,high risk +33,120,75,10,98,70,high risk +48,120,80,11,98,88,high risk +50,140,95,17,98,60,high risk +30,140,100,15,98,70,high risk +29,120,75,7.2,100,70,high risk +48,120,80,11,98,88,high risk +50,140,90,15,98,77,high risk +25,140,100,7.2,98,80,high risk +55,140,80,7.2,101,76,high risk +40,140,100,18,98,77,high risk +28,120,80,9,102,76,high risk +32,140,100,8,98,70,high risk +17,90,60,11,101,78,high risk +17,90,63,8,101,70,high risk +25,120,90,12,101,80,high risk +17,120,80,7,102,76,high risk +19,90,65,11,101,70,high risk +37,120,90,11,98,88,high risk +17,110,75,13,101,76,high risk +25,120,90,15,98,80,high risk +40,160,100,19,98,77,high risk +32,140,90,18,98,88,high risk +12,90,60,8,102,66,high risk +35,140,100,9,98,66,high risk +54,140,100,15,98,66,high risk +40,120,95,11,98,80,high risk +12,90,60,11,102,60,high risk +60,120,85,15,98,60,high risk +55,140,95,19,98,77,high risk +50,130,100,16,98,76,high risk +13,90,65,9,101,80,high risk +17,90,65,7.7,103,67,high risk +17,85,60,6.3,102,86,high risk +40,120,75,7.7,98,70,high risk +48,120,80,11,98,88,high risk +22,120,60,15,98,80,high risk +55,120,90,18,98,60,high risk +35,85,60,19,98,86,high risk +43,120,90,18,98,70,high risk +32,120,65,6,101,76,mid risk diff --git a/Maternal health Risk Prediction/app.py b/Maternal health Risk Prediction/app.py new file mode 100644 index 00000000..513afb11 --- /dev/null +++ b/Maternal health Risk Prediction/app.py @@ -0,0 +1,117 @@ +import streamlit as st +import joblib +import pandas as pd + +# Load the trained XGBoost model +model_xgb = joblib.load('xgb_model.pkl') + +# Function to make predictions +def predict_risk_level(age, systolic_bp, diastolic_bp, bs, body_temp, heart_rate): + # Create a DataFrame for the input data + input_data = pd.DataFrame({ + 'Age': [age], + 'SystolicBP': [systolic_bp], + 'DiastolicBP': [diastolic_bp], + 'BS': [bs], + 'BodyTemp': [body_temp], + 'HeartRate': [heart_rate] + }) + + # Predict using the loaded model + prediction_proba = model_xgb.predict_proba(input_data)[0] + + # Determine risk level based on probability thresholds + low_risk_threshold = 0.33 + mid_risk_threshold = 0.66 + + if prediction_proba[2] > mid_risk_threshold: + risk_level = 'High Maternal risk' + elif prediction_proba[1] > low_risk_threshold: + risk_level = 'Medium Maternal risk' + else: + risk_level = 'Low Maternal risk' + + return risk_level + +# Streamlit app interface +st.set_page_config(page_title="Maternal Risk Prediction", page_icon=":baby:", layout="wide") + +# Add a maternal-themed image as background with blur +st.markdown(""" + + """, unsafe_allow_html=True) + +# Background image +st.markdown("
", unsafe_allow_html=True) + +# Main title +st.markdown("
Maternal Risk Prediction
", unsafe_allow_html=True) + + +# Input container with styling +with st.container(): + st.markdown("
Enter the details below:
", unsafe_allow_html=True) + + with st.container(): + col1, col2 = st.columns(2) + + with col1: + age = st.slider('Age', min_value=10, max_value=100, value=30, key='age') + systolic_bp = st.slider('Systolic BP', min_value=70, max_value=200, value=120, key='systolic_bp') + diastolic_bp = st.slider('Diastolic BP', min_value=50, max_value=120, value=80, key='diastolic_bp') + bs = st.slider('Blood Sugar', min_value=5.0, max_value=20.0, value=7.0, format="%.1f", key='bs') + + with col2: + body_temp = st.slider('Body Temperature (Fahrenheit)', min_value=95.0, max_value=105.0, value=98.6, + format="%.1f", key='body_temp') + heart_rate = st.slider('Heart Rate', min_value=50, max_value=150, value=80, key='heart_rate') + + st.markdown("
", unsafe_allow_html=True) + + # Prediction button + if st.button('Predict Risk Level'): + risk_level = predict_risk_level(age, systolic_bp, diastolic_bp, bs, body_temp, heart_rate) + + # Display the prediction with proper styling + with st.container(): + if risk_level == 'Low Maternal risk': + st.success(f'**Predicted Risk Level:** {risk_level.upper()}') + elif risk_level == 'Medium Maternal risk': + st.warning(f'**Predicted Risk Level:** {risk_level.upper()}') + elif risk_level == 'High Maternal risk': + st.error(f'**Predicted Risk Level:** {risk_level.upper()}') diff --git a/Maternal health Risk Prediction/images/bar plot of risk level after oversampling.png b/Maternal health Risk Prediction/images/bar plot of risk level after oversampling.png new file mode 100644 index 00000000..1428b65e Binary files /dev/null and b/Maternal health Risk Prediction/images/bar plot of risk level after oversampling.png differ diff --git a/Maternal health Risk Prediction/images/bar plot of risk level.png b/Maternal health Risk Prediction/images/bar plot of risk level.png new file mode 100644 index 00000000..8d640edc Binary files /dev/null and b/Maternal health Risk Prediction/images/bar plot of risk level.png differ diff --git a/Maternal health Risk Prediction/images/box plot of systolicBP by risk level.png b/Maternal health Risk Prediction/images/box plot of systolicBP by risk level.png new file mode 100644 index 00000000..819d4aee Binary files /dev/null and b/Maternal health Risk Prediction/images/box plot of systolicBP by risk level.png differ diff --git a/Maternal health Risk Prediction/images/box_plot.png b/Maternal health Risk Prediction/images/box_plot.png new file mode 100644 index 00000000..c4f0726b Binary files /dev/null and b/Maternal health Risk Prediction/images/box_plot.png differ diff --git a/Maternal health Risk Prediction/images/box_plot_unprocessed.png b/Maternal health Risk Prediction/images/box_plot_unprocessed.png new file mode 100644 index 00000000..634c30ca Binary files /dev/null and b/Maternal health Risk Prediction/images/box_plot_unprocessed.png differ diff --git a/Maternal health Risk Prediction/images/cm_randomforest.png b/Maternal health Risk Prediction/images/cm_randomforest.png new file mode 100644 index 00000000..1b8acb5d Binary files /dev/null and b/Maternal health Risk Prediction/images/cm_randomforest.png differ diff --git a/Maternal health Risk Prediction/images/confusion_matrix_gbm.png b/Maternal health Risk Prediction/images/confusion_matrix_gbm.png new file mode 100644 index 00000000..64a35035 Binary files /dev/null and b/Maternal health Risk Prediction/images/confusion_matrix_gbm.png differ diff --git a/Maternal health Risk Prediction/images/confusion_matrix_knn.png b/Maternal health Risk Prediction/images/confusion_matrix_knn.png new file mode 100644 index 00000000..e8640230 Binary files /dev/null and b/Maternal health Risk Prediction/images/confusion_matrix_knn.png differ diff --git a/Maternal health Risk Prediction/images/confusion_matrix_xgboost.png b/Maternal health Risk Prediction/images/confusion_matrix_xgboost.png new file mode 100644 index 00000000..5e64f4a2 Binary files /dev/null and b/Maternal health Risk Prediction/images/confusion_matrix_xgboost.png differ diff --git a/Maternal health Risk Prediction/images/correlation heatmap.png b/Maternal health Risk Prediction/images/correlation heatmap.png new file mode 100644 index 00000000..de9eeb73 Binary files /dev/null and b/Maternal health Risk Prediction/images/correlation heatmap.png differ diff --git a/Maternal health Risk Prediction/images/histogram of age.png b/Maternal health Risk Prediction/images/histogram of age.png new file mode 100644 index 00000000..c6cd8389 Binary files /dev/null and b/Maternal health Risk Prediction/images/histogram of age.png differ diff --git a/Maternal health Risk Prediction/images/line plot of avg BS by age.png b/Maternal health Risk Prediction/images/line plot of avg BS by age.png new file mode 100644 index 00000000..195b4e3f Binary files /dev/null and b/Maternal health Risk Prediction/images/line plot of avg BS by age.png differ diff --git a/Maternal health Risk Prediction/images/model_comparison.png b/Maternal health Risk Prediction/images/model_comparison.png new file mode 100644 index 00000000..505c6240 Binary files /dev/null and b/Maternal health Risk Prediction/images/model_comparison.png differ diff --git a/Maternal health Risk Prediction/images/scatter_plot_BSvsAge.png b/Maternal health Risk Prediction/images/scatter_plot_BSvsAge.png new file mode 100644 index 00000000..68eb4a49 Binary files /dev/null and b/Maternal health Risk Prediction/images/scatter_plot_BSvsAge.png differ diff --git a/Maternal health Risk Prediction/images/scatter_plot_SBPVsDBP.png b/Maternal health Risk Prediction/images/scatter_plot_SBPVsDBP.png new file mode 100644 index 00000000..f5b31074 Binary files /dev/null and b/Maternal health Risk Prediction/images/scatter_plot_SBPVsDBP.png differ diff --git a/Maternal health Risk Prediction/images/violin_plot_of_heart_ratevs Risk level.png b/Maternal health Risk Prediction/images/violin_plot_of_heart_ratevs Risk level.png new file mode 100644 index 00000000..88e49481 Binary files /dev/null and b/Maternal health Risk Prediction/images/violin_plot_of_heart_ratevs Risk level.png differ diff --git a/Maternal health Risk Prediction/maternal_health_risk_prediction.ipynb b/Maternal health Risk Prediction/maternal_health_risk_prediction.ipynb new file mode 100644 index 00000000..d47ebdf9 --- /dev/null +++ b/Maternal health Risk Prediction/maternal_health_risk_prediction.ipynb @@ -0,0 +1,2519 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "provenance": [] + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "language_info": { + "name": "python" + } + }, + "cells": [ + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "kqZhgJMut--e" + }, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "import warnings\n", + "warnings.filterwarnings('ignore')\n", + "import plotly.express as px\n", + "import plotly.graph_objects as go" + ] + }, + { + "cell_type": "code", + "source": [ + "df=pd.read_csv('/content/Maternal Health Risk Data Set.csv')" + ], + "metadata": { + "id": "3ctp6U8xuRIp" + }, + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "df.head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "EGM-fhIvuXjf", + "outputId": "cb0ffce8-e24b-4d7a-dd47-6e2cb6589b43" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Age SystolicBP DiastolicBP BS BodyTemp HeartRate RiskLevel\n", + "0 25 130 80 15.0 98.0 86 high risk\n", + "1 35 140 90 13.0 98.0 70 high risk\n", + "2 29 90 70 8.0 100.0 80 high risk\n", + "3 30 140 85 7.0 98.0 70 high risk\n", + "4 35 120 60 6.1 98.0 76 low risk" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
AgeSystolicBPDiastolicBPBSBodyTempHeartRateRiskLevel
0251308015.098.086high risk
1351409013.098.070high risk
22990708.0100.080high risk
330140857.098.070high risk
435120606.198.076low risk
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "df", + "summary": "{\n \"name\": \"df\",\n \"rows\": 1014,\n \"fields\": [\n {\n \"column\": \"Age\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 13,\n \"min\": 10,\n \"max\": 70,\n \"num_unique_values\": 50,\n \"samples\": [\n 40,\n 43,\n 13\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"SystolicBP\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 18,\n \"min\": 70,\n \"max\": 160,\n \"num_unique_values\": 19,\n \"samples\": [\n 130,\n 110,\n 80\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"DiastolicBP\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 13,\n \"min\": 49,\n \"max\": 100,\n \"num_unique_values\": 16,\n \"samples\": [\n 80,\n 90,\n 89\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"BS\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 3.293531721151281,\n \"min\": 6.0,\n \"max\": 19.0,\n \"num_unique_values\": 29,\n \"samples\": [\n 6.5,\n 7.7,\n 7.1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"BodyTemp\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.3713843755995376,\n \"min\": 98.0,\n \"max\": 103.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 100.0,\n 98.4,\n 98.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"HeartRate\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8,\n \"min\": 7,\n \"max\": 90,\n \"num_unique_values\": 16,\n \"samples\": [\n 86,\n 70,\n 77\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"RiskLevel\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"high risk\",\n \"low risk\",\n \"mid risk\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 23 + } + ] + }, + { + "cell_type": "code", + "source": [ + "df.info()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "xrtLbXFmuaQH", + "outputId": "48925481-5e99-40bd-aa91-56e935c81592" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\n", + "RangeIndex: 1014 entries, 0 to 1013\n", + "Data columns (total 7 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Age 1014 non-null int64 \n", + " 1 SystolicBP 1014 non-null int64 \n", + " 2 DiastolicBP 1014 non-null int64 \n", + " 3 BS 1014 non-null float64\n", + " 4 BodyTemp 1014 non-null float64\n", + " 5 HeartRate 1014 non-null int64 \n", + " 6 RiskLevel 1014 non-null object \n", + "dtypes: float64(2), int64(4), object(1)\n", + "memory usage: 55.6+ KB\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "df.describe()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 300 + }, + "id": "0t-G0R62udqd", + "outputId": "3f18771e-79eb-43f2-f8f1-c402cf6a0c0e" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Age SystolicBP DiastolicBP BS BodyTemp \\\n", + "count 1014.000000 1014.000000 1014.000000 1014.000000 1014.000000 \n", + "mean 29.871795 113.198225 76.460552 8.725986 98.665089 \n", + "std 13.474386 18.403913 13.885796 3.293532 1.371384 \n", + "min 10.000000 70.000000 49.000000 6.000000 98.000000 \n", + "25% 19.000000 100.000000 65.000000 6.900000 98.000000 \n", + "50% 26.000000 120.000000 80.000000 7.500000 98.000000 \n", + "75% 39.000000 120.000000 90.000000 8.000000 98.000000 \n", + "max 70.000000 160.000000 100.000000 19.000000 103.000000 \n", + "\n", + " HeartRate \n", + "count 1014.000000 \n", + "mean 74.301775 \n", + "std 8.088702 \n", + "min 7.000000 \n", + "25% 70.000000 \n", + "50% 76.000000 \n", + "75% 80.000000 \n", + "max 90.000000 " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
AgeSystolicBPDiastolicBPBSBodyTempHeartRate
count1014.0000001014.0000001014.0000001014.0000001014.0000001014.000000
mean29.871795113.19822576.4605528.72598698.66508974.301775
std13.47438618.40391313.8857963.2935321.3713848.088702
min10.00000070.00000049.0000006.00000098.0000007.000000
25%19.000000100.00000065.0000006.90000098.00000070.000000
50%26.000000120.00000080.0000007.50000098.00000076.000000
75%39.000000120.00000090.0000008.00000098.00000080.000000
max70.000000160.000000100.00000019.000000103.00000090.000000
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"df\",\n \"rows\": 8,\n \"fields\": [\n {\n \"column\": \"Age\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 348.54126044861886,\n \"min\": 10.0,\n \"max\": 1014.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 29.871794871794872,\n 26.0,\n 1014.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"SystolicBP\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 325.7381080591285,\n \"min\": 18.403912756342706,\n \"max\": 1014.0,\n \"num_unique_values\": 7,\n \"samples\": [\n 1014.0,\n 113.19822485207101,\n 120.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"DiastolicBP\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 335.61546340338174,\n \"min\": 13.885795724160687,\n \"max\": 1014.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 76.46055226824457,\n 80.0,\n 1014.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"BS\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 355.5316275877343,\n \"min\": 3.293531721151281,\n \"max\": 1014.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 8.725986193293886,\n 7.5,\n 1014.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"BodyTemp\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 330.2234922746751,\n \"min\": 1.3713843755995376,\n \"max\": 1014.0,\n \"num_unique_values\": 5,\n \"samples\": [\n 98.66508875739645,\n 103.0,\n 1.3713843755995376\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"HeartRate\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 339.57400097574833,\n \"min\": 7.0,\n \"max\": 1014.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 74.30177514792899,\n 76.0,\n 1014.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 25 + } + ] + }, + { + "cell_type": "code", + "source": [ + "df.isna().sum()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "i3bRp6SLugPc", + "outputId": "fd958808-59b4-4ec0-da37-f6d1c585995c" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Age 0\n", + "SystolicBP 0\n", + "DiastolicBP 0\n", + "BS 0\n", + "BodyTemp 0\n", + "HeartRate 0\n", + "RiskLevel 0\n", + "dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 26 + } + ] + }, + { + "cell_type": "code", + "source": [ + "df.shape" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "pl6dOvQNujF6", + "outputId": "2c6b98b9-83bd-42c0-c422-e05b4c3b9c90" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(1014, 7)" + ] + }, + "metadata": {}, + "execution_count": 27 + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Printing unique values in each column\n", + "for column in df.columns:\n", + " unique_values = df[column].unique()\n", + " print(f\"Unique values in '{column}': {unique_values}\")" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "NzkKeD0Ju1XD", + "outputId": "7156f24b-6d62-431d-918c-6f6f9c411d0b" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Unique values in 'Age': [25 35 29 30 23 32 42 19 20 48 15 50 10 40 21 18 16 22 49 28 12 60 55 45\n", + " 31 17 26 54 44 33 13 34 38 39 63 14 37 51 62 43 65 66 56 70 27 36 59 24\n", + " 41 46]\n", + "Unique values in 'SystolicBP': [130 140 90 120 85 110 70 100 75 95 76 80 115 135 160 129 83 99\n", + " 78]\n", + "Unique values in 'DiastolicBP': [ 80 90 70 85 60 89 75 100 50 65 95 49 63 69 76 68]\n", + "Unique values in 'BS': [15. 13. 8. 7. 6.1 7.01 11. 6.9 18. 6.7 7.5 7.2\n", + " 7.1 6.4 9. 6. 7.7 12. 16. 7.8 6.8 7.9 17. 19.\n", + " 10. 6.3 6.6 6.5 7.6 ]\n", + "Unique values in 'BodyTemp': [ 98. 100. 102. 101. 103. 98.4 99. 98.6]\n", + "Unique values in 'HeartRate': [86 70 80 76 78 77 88 90 66 82 60 75 67 65 68 7]\n", + "Unique values in 'RiskLevel': ['high risk' 'low risk' 'mid risk']\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "# 1. Histogram of Age\n", + "fig1 = px.histogram(df, x=\"Age\", title=\"Histogram of Age\")\n", + "fig1.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 542 + }, + "id": "GE6G6oZOu4Sw", + "outputId": "49b75c20-f29f-4fd4-a093-5a0ee1c3852a" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# 2. Box Plot of SystolicBP by RiskLevel\n", + "fig2 = px.box(df, x=\"RiskLevel\", y=\"SystolicBP\", title=\"Box Plot of SystolicBP by Risk Level\")\n", + "fig2.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 542 + }, + "id": "Bec2efz1vZQF", + "outputId": "3a8c797f-bd66-4efd-93b3-67d1994cf2fe" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# 3. Scatter Plot of SystolicBP vs DiastolicBP\n", + "fig3 = px.scatter(df, x=\"SystolicBP\", y=\"DiastolicBP\", color=\"RiskLevel\", title=\"Scatter Plot of SystolicBP vs DiastolicBP\")\n", + "fig3.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 542 + }, + "id": "SxjjsLt6vd9Z", + "outputId": "07b9dda5-cd9f-4733-96d1-65231c8a7b9c" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# 4. Scatter Plot of BS vs Age\n", + "fig4 = px.scatter(df, x=\"Age\", y=\"BS\", color=\"RiskLevel\", title=\"Scatter Plot of BS vs Age\")\n", + "fig4.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 542 + }, + "id": "Yi2EKDZBvhg-", + "outputId": "081ae2a8-aa73-4d7e-8d07-247a9078931c" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# 5. Violin Plot of HeartRate by RiskLevel\n", + "fig5 = px.violin(df, y=\"HeartRate\", x=\"RiskLevel\", box=True, points=\"all\", title=\"Violin Plot of Heart Rate by Risk Level\")\n", + "fig5.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 542 + }, + "id": "w7osC9w-vmNa", + "outputId": "b62eb498-4b64-4623-fe45-2710ebb43d3c" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# 6. Line Plot of Average BS by Age\n", + "avg_bs_by_age = df.groupby('Age')['BS'].mean().reset_index()\n", + "fig7 = px.line(avg_bs_by_age, x='Age', y='BS', title=\"Line Plot of Average BS by Age\")\n", + "fig7.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 542 + }, + "id": "nFpFXOPKvqQL", + "outputId": "bb3ba357-30ea-4e32-a1e7-53f5c3fa6293" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# 7. Bar Plot of RiskLevel\n", + "risk_counts = df['RiskLevel'].value_counts().reset_index()\n", + "risk_counts.columns = ['RiskLevel', 'count']\n", + "fig7 = px.bar(risk_counts, x='RiskLevel', y='count', title=\"Bar Plot of Risk Level\")\n", + "fig7.update_xaxes(title=\"Risk Level\")\n", + "fig7.update_yaxes(title=\"Count\")\n", + "fig7.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 542 + }, + "id": "L0F-a-bhv4lj", + "outputId": "9a32f418-d3d3-4ba7-a7b4-f78987e7d9ca" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Print the counts\n", + "print(risk_counts)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "Flou_5gow0_9", + "outputId": "adf0d8d4-da42-484e-e65c-92544ecb8138" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " RiskLevel count\n", + "0 low risk 406\n", + "1 mid risk 336\n", + "2 high risk 272\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "from imblearn.over_sampling import SMOTE\n", + "from sklearn.preprocessing import LabelEncoder\n", + "\n", + "# Encode the 'RiskLevel' column\n", + "le = LabelEncoder()\n", + "df['RiskLevel'] = le.fit_transform(df['RiskLevel'])\n", + "\n", + "# Separate features and target\n", + "X = df.drop(columns=['RiskLevel'])\n", + "y = df['RiskLevel']\n", + "\n", + "# Apply SMOTE\n", + "smote = SMOTE()\n", + "X_resampled, y_resampled = smote.fit_resample(X, y)\n", + "\n", + "# Decode the resampled target\n", + "df = X_resampled.copy()\n", + "df['RiskLevel'] = le.inverse_transform(y_resampled)\n", + "\n", + "# Count the unique values in 'RiskLevel' after oversampling\n", + "risk_counts_resampled = df['RiskLevel'].value_counts().reset_index()\n", + "risk_counts_resampled.columns = ['RiskLevel', 'count']\n", + "\n", + "# Print the counts\n", + "print(risk_counts_resampled)\n", + "\n", + "# 6. Bar Plot of RiskLevel after oversampling\n", + "fig6 = px.bar(risk_counts_resampled, x='RiskLevel', y='count', title=\"Bar Plot of Risk Level After Oversampling\")\n", + "fig6.update_xaxes(title=\"Risk Level\")\n", + "fig6.update_yaxes(title=\"Count\")\n", + "fig6.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 611 + }, + "id": "EReSNRtSw91f", + "outputId": "5a889143-417e-4323-f00a-e8e92752d7a2" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " RiskLevel count\n", + "0 high risk 406\n", + "1 low risk 406\n", + "2 mid risk 406\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "
\n", + "
\n", + "\n", + "" + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Encode the 'RiskLevel' column\n", + "le = LabelEncoder()\n", + "df['RiskLevel'] = le.fit_transform(df['RiskLevel'])" + ], + "metadata": { + "id": "6_wWpUVLyygZ" + }, + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "df.head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "nTtRgFI8gw-_", + "outputId": "7ba229a4-6bc3-4042-e939-395a0c033a9f" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Age SystolicBP DiastolicBP BS BodyTemp HeartRate RiskLevel\n", + "0 25 130 80 15.0 98.0 86 0\n", + "1 35 140 90 13.0 98.0 70 0\n", + "2 29 90 70 8.0 100.0 80 0\n", + "3 30 140 85 7.0 98.0 70 0\n", + "4 35 120 60 6.1 98.0 76 1" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
AgeSystolicBPDiastolicBPBSBodyTempHeartRateRiskLevel
0251308015.098.0860
1351409013.098.0700
22990708.0100.0800
330140857.098.0700
435120606.198.0761
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "df", + "summary": "{\n \"name\": \"df\",\n \"rows\": 1218,\n \"fields\": [\n {\n \"column\": \"Age\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 13,\n \"min\": 10,\n \"max\": 70,\n \"num_unique_values\": 54,\n \"samples\": [\n 28,\n 46,\n 41\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"SystolicBP\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 18,\n \"min\": 70,\n \"max\": 160,\n \"num_unique_values\": 30,\n \"samples\": [\n 122,\n 129,\n 128\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"DiastolicBP\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14,\n \"min\": 49,\n \"max\": 100,\n \"num_unique_values\": 29,\n \"samples\": [\n 91,\n 62,\n 63\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"BS\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 3.5563601902629047,\n \"min\": 6.0,\n \"max\": 19.0,\n \"num_unique_values\": 114,\n \"samples\": [\n 8.005702919151267,\n 6.1,\n 7.6525385267807815\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"BodyTemp\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.396702331767838,\n \"min\": 98.0,\n \"max\": 103.0,\n \"num_unique_values\": 40,\n \"samples\": [\n 99.62208840265713,\n 100.23717269342228,\n 98.23667547310149\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"HeartRate\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8,\n \"min\": 7,\n \"max\": 90,\n \"num_unique_values\": 24,\n \"samples\": [\n 66,\n 73,\n 86\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"RiskLevel\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 2,\n \"num_unique_values\": 3,\n \"samples\": [\n 0,\n 1,\n 2\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 39 + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Plot the initial box plots\n", + "plt.figure(figsize=(15, 10))\n", + "for i, column in enumerate(df.columns[:-1], 1):\n", + " plt.subplot(2, 3, i)\n", + " sns.boxplot(y=df[column])\n", + " plt.title(column)\n", + "plt.tight_layout()\n", + "plt.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 691 + }, + "id": "gWhuE_aKh1ee", + "outputId": "c39c3bfa-4a8e-422f-e176-2ef7ea7caac7" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Function to remove outliers using IQR\n", + "def remove_outliers(df, column):\n", + " Q1 = df[column].quantile(0.25)\n", + " Q3 = df[column].quantile(0.75)\n", + " IQR = Q3 - Q1\n", + " lower_bound = Q1 - 1.5 * IQR\n", + " upper_bound = Q3 + 1.5 * IQR\n", + " return df[(df[column] >= lower_bound) & (df[column] <= upper_bound)]\n", + "\n", + "# Remove outliers from each column\n", + "df_no_outliers = df.copy()\n", + "for column in df.columns[:-1]:\n", + " df_no_outliers = remove_outliers(df_no_outliers, column)\n", + "\n", + "# Plot the box plots again after removing outliers\n", + "plt.figure(figsize=(15, 10))\n", + "for i, column in enumerate(df_no_outliers.columns[:-1], 1):\n", + " plt.subplot(2, 3, i)\n", + " sns.boxplot(y=df_no_outliers[column])\n", + " plt.title(column)\n", + "plt.tight_layout()\n", + "plt.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 691 + }, + "id": "FNVlqy6WhqQz", + "outputId": "079f129b-2348-4295-f2ac-f3d96430a1d9" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Calculate the correlation matrix\n", + "corr_matrix = df.corr()\n", + "\n", + "# Plot the correlation matrix as a heatmap\n", + "plt.figure(figsize=(10, 8))\n", + "sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', fmt='.2f', linewidths=0.5)\n", + "plt.title('Correlation Matrix')\n", + "plt.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 699 + }, + "id": "dI7THy6miaQ2", + "outputId": "95836805-2676-41e5-aac4-a2b9a6e3bb32" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# **Random Forest**" + ], + "metadata": { + "id": "D9BH1CrSj376" + } + }, + { + "cell_type": "code", + "source": [ + "from sklearn.model_selection import train_test_split\n", + "from sklearn.ensemble import RandomForestClassifier\n", + "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, classification_report, confusion_matrix\n", + "\n", + "# Features and target\n", + "X = df.drop('RiskLevel', axis=1)\n", + "y = df['RiskLevel']\n", + "\n", + "# Split the data\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", + "\n", + "# Train a Random Forest classifier\n", + "model_rf = RandomForestClassifier(random_state=42)\n", + "model_rf.fit(X_train, y_train)\n", + "\n", + "# Make predictions\n", + "y_pred = model_rf.predict(X_test)\n", + "\n", + "# Calculate metrics\n", + "accuracy_rf = accuracy_score(y_test, y_pred)\n", + "precision_rf = precision_score(y_test, y_pred, average='weighted')\n", + "recall_rf = recall_score(y_test, y_pred, average='weighted')\n", + "f1_score_rf = f1_score(y_test, y_pred, average='weighted')\n", + "classification_report_rf = classification_report(y_test, y_pred)\n", + "confusion_matrix_rf = confusion_matrix(y_test, y_pred)\n", + "\n", + "# Output the results\n", + "print(f\"Accuracy: {accuracy_rf}\")\n", + "print(f\"Precision: {precision_rf}\")\n", + "print(f\"Recall: {recall_rf}\")\n", + "print(f\"F1 Score: {f1_score_rf}\")\n", + "print(\"Classification Report:\")\n", + "print(classification_report_rf)\n", + "print(\"Confusion Matrix:\")\n", + "print(confusion_matrix_rf)\n", + "\n", + "# Plot the confusion matrix using Seaborn\n", + "plt.figure(figsize=(10, 7))\n", + "sns.heatmap(confusion_matrix_rf, annot=True, fmt='d', cmap='Blues', xticklabels=['Low Risk', 'Mid Risk', 'High Risk'], yticklabels=['Low Risk', 'Mid Risk', 'High Risk'])\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Actual')\n", + "plt.title('Confusion Matrix')\n", + "plt.show()\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 971 + }, + "id": "b-C_UpNvhqNc", + "outputId": "f8313b82-eb33-42a4-991d-d23271af2959" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy: 0.8565573770491803\n", + "Precision: 0.8612761809891025\n", + "Recall: 0.8565573770491803\n", + "F1 Score: 0.8579504862964055\n", + "Classification Report:\n", + " precision recall f1-score support\n", + "\n", + " 0 1.00 0.95 0.97 73\n", + " 1 0.83 0.79 0.81 87\n", + " 2 0.77 0.85 0.81 84\n", + "\n", + " accuracy 0.86 244\n", + " macro avg 0.87 0.86 0.86 244\n", + "weighted avg 0.86 0.86 0.86 244\n", + "\n", + "Confusion Matrix:\n", + "[[69 1 3]\n", + " [ 0 69 18]\n", + " [ 0 13 71]]\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# **XGBOOST**" + ], + "metadata": { + "id": "kfBqf3tpjzie" + } + }, + { + "cell_type": "code", + "source": [ + "from xgboost import XGBClassifier\n", + "\n", + "# Train an XGBoost classifier\n", + "model_xgb = XGBClassifier(random_state=42, use_label_encoder=False, eval_metric='logloss')\n", + "model_xgb.fit(X_train, y_train)\n", + "\n", + "# Make predictions\n", + "y_pred_xgb = model_xgb.predict(X_test)\n", + "\n", + "# Calculate metrics\n", + "accuracy_xgb = accuracy_score(y_test, y_pred_xgb)\n", + "precision_xgb = precision_score(y_test, y_pred_xgb, average='weighted')\n", + "recall_xgb = recall_score(y_test, y_pred_xgb, average='weighted')\n", + "f1_score_xgb = f1_score(y_test, y_pred_xgb, average='weighted')\n", + "classification_report_xgb = classification_report(y_test, y_pred_xgb)\n", + "confusion_matrix_xgb = confusion_matrix(y_test, y_pred_xgb)\n", + "\n", + "# Output the results\n", + "print(f\"Accuracy: {accuracy_xgb}\")\n", + "print(f\"Precision: {precision_xgb}\")\n", + "print(f\"Recall: {recall_xgb}\")\n", + "print(f\"F1 Score: {f1_score_xgb}\")\n", + "print(\"Classification Report:\")\n", + "print(classification_report_xgb)\n", + "print(\"Confusion Matrix:\")\n", + "print(confusion_matrix_xgb)\n", + "\n", + "# Plot the confusion matrix using Seaborn\n", + "plt.figure(figsize=(10, 7))\n", + "sns.heatmap(confusion_matrix_xgb, annot=True, fmt='d', cmap='Blues', xticklabels=['Low Risk', 'Mid Risk', 'High Risk'], yticklabels=['Low Risk', 'Mid Risk', 'High Risk'])\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Actual')\n", + "plt.title('Confusion Matrix')\n", + "plt.show()\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 971 + }, + "id": "p0EeBTi-jsSh", + "outputId": "89170f6c-ba1b-4dd0-860c-5846b93edeb9" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy: 0.8647540983606558\n", + "Precision: 0.8678961748633879\n", + "Recall: 0.8647540983606558\n", + "F1 Score: 0.8657718269750044\n", + "Classification Report:\n", + " precision recall f1-score support\n", + "\n", + " 0 1.00 0.96 0.98 73\n", + " 1 0.83 0.80 0.82 87\n", + " 2 0.79 0.85 0.82 84\n", + "\n", + " accuracy 0.86 244\n", + " macro avg 0.87 0.87 0.87 244\n", + "weighted avg 0.87 0.86 0.87 244\n", + "\n", + "Confusion Matrix:\n", + "[[70 1 2]\n", + " [ 0 70 17]\n", + " [ 0 13 71]]\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# **KNN**" + ], + "metadata": { + "id": "edjeeuXWkYn5" + } + }, + { + "cell_type": "code", + "source": [ + "from sklearn.neighbors import KNeighborsClassifier\n", + "\n", + "# Train a KNN classifier\n", + "model_knn = KNeighborsClassifier()\n", + "model_knn.fit(X_train, y_train)\n", + "\n", + "# Make predictions\n", + "y_pred_knn = model_knn.predict(X_test)\n", + "\n", + "# Calculate metrics\n", + "accuracy_knn = accuracy_score(y_test, y_pred_knn)\n", + "precision_knn = precision_score(y_test, y_pred_knn, average='weighted')\n", + "recall_knn = recall_score(y_test, y_pred_knn, average='weighted')\n", + "f1_score_knn = f1_score(y_test, y_pred_knn, average='weighted')\n", + "classification_report_knn = classification_report(y_test, y_pred_knn)\n", + "confusion_matrix_knn = confusion_matrix(y_test, y_pred_knn)\n", + "\n", + "# Output the results\n", + "print(f\"Accuracy: {accuracy_knn}\")\n", + "print(f\"Precision: {precision_knn}\")\n", + "print(f\"Recall: {recall_knn}\")\n", + "print(f\"F1 Score: {f1_score_knn}\")\n", + "print(\"Classification Report:\")\n", + "print(classification_report_knn)\n", + "print(\"Confusion Matrix:\")\n", + "print(confusion_matrix_knn)\n", + "\n", + "# Plot the confusion matrix using Seaborn\n", + "plt.figure(figsize=(10, 7))\n", + "sns.heatmap(confusion_matrix_knn, annot=True, fmt='d', cmap='Blues', xticklabels=['Low Risk', 'Mid Risk', 'High Risk'], yticklabels=['Low Risk', 'Mid Risk', 'High Risk'])\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Actual')\n", + "plt.title('Confusion Matrix')\n", + "plt.show()\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 971 + }, + "id": "ZAl2U0FQkQJd", + "outputId": "109d13d0-fe47-4233-f234-75d3873c7efa" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy: 0.7745901639344263\n", + "Precision: 0.7821106043050179\n", + "Recall: 0.7745901639344263\n", + "F1 Score: 0.7766816406556944\n", + "Classification Report:\n", + " precision recall f1-score support\n", + "\n", + " 0 0.94 0.86 0.90 73\n", + " 1 0.74 0.70 0.72 87\n", + " 2 0.68 0.77 0.73 84\n", + "\n", + " accuracy 0.77 244\n", + " macro avg 0.79 0.78 0.78 244\n", + "weighted avg 0.78 0.77 0.78 244\n", + "\n", + "Confusion Matrix:\n", + "[[63 3 7]\n", + " [ 3 61 23]\n", + " [ 1 18 65]]\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# **GBM**" + ], + "metadata": { + "id": "QAzCcnKbkwJv" + } + }, + { + "cell_type": "code", + "source": [ + "from sklearn.ensemble import GradientBoostingClassifier\n", + "\n", + "# Train a GBM classifier\n", + "model_gbm = GradientBoostingClassifier(random_state=42)\n", + "model_gbm.fit(X_train, y_train)\n", + "\n", + "# Make predictions\n", + "y_pred_gbm = model_gbm.predict(X_test)\n", + "\n", + "# Calculate metrics\n", + "accuracy_gbm = accuracy_score(y_test, y_pred_gbm)\n", + "precision_gbm = precision_score(y_test, y_pred_gbm, average='weighted')\n", + "recall_gbm = recall_score(y_test, y_pred_gbm, average='weighted')\n", + "f1_score_gbm = f1_score(y_test, y_pred_gbm, average='weighted')\n", + "classification_report_gbm = classification_report(y_test, y_pred_gbm)\n", + "confusion_matrix_gbm = confusion_matrix(y_test, y_pred_gbm)\n", + "\n", + "# Output the results\n", + "print(f\"Accuracy: {accuracy_gbm}\")\n", + "print(f\"Precision: {precision_gbm}\")\n", + "print(f\"Recall: {recall_gbm}\")\n", + "print(f\"F1 Score: {f1_score_gbm}\")\n", + "print(\"Classification Report:\")\n", + "print(classification_report_gbm)\n", + "print(\"Confusion Matrix:\")\n", + "print(confusion_matrix_gbm)\n", + "\n", + "# Plot the confusion matrix using Seaborn\n", + "plt.figure(figsize=(10, 7))\n", + "sns.heatmap(confusion_matrix_gbm, annot=True, fmt='d', cmap='Blues', xticklabels=['Low Risk', 'Mid Risk', 'High Risk'], yticklabels=['Low Risk', 'Mid Risk', 'High Risk'])\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Actual')\n", + "plt.title('Confusion Matrix')\n", + "plt.show()\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 971 + }, + "id": "K9MmYOxSkrNg", + "outputId": "cfd211d9-93a2-48b8-d546-088499dd2c4d" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy: 0.7909836065573771\n", + "Precision: 0.8054644808743169\n", + "Recall: 0.7909836065573771\n", + "F1 Score: 0.7944936110768726\n", + "Classification Report:\n", + " precision recall f1-score support\n", + "\n", + " 0 0.98 0.81 0.89 73\n", + " 1 0.75 0.76 0.75 87\n", + " 2 0.71 0.81 0.76 84\n", + "\n", + " accuracy 0.79 244\n", + " macro avg 0.81 0.79 0.80 244\n", + "weighted avg 0.81 0.79 0.79 244\n", + "\n", + "Confusion Matrix:\n", + "[[59 6 8]\n", + " [ 1 66 20]\n", + " [ 0 16 68]]\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwgAAAJwCAYAAAAtA0YPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABdZklEQVR4nO3deXzM1xrH8e8ksiG7JShBkaB2ilpCG1VVRawttdPFUtIFrdpJq7XW1laIq5baW6qWUlSpqqUoQi2NLfallgTJ3D/CmGlEMxr5jc7nfV/zesn5nfmdZ3IvN0+e8/yOyWw2mwUAAAAAklyMDgAAAACA4yBBAAAAAGBBggAAAADAggQBAAAAgAUJAgAAAAALEgQAAAAAFiQIAAAAACxIEAAAAABYkCAAAAAAsCBBAIB7OHDggJ599ln5+vrKZDJp8eLFGXr/I0eOyGQyKSYmJkPv+yirVauWatWqZXQYAOD0SBAAOKyDBw/q1VdfVeHCheXp6SkfHx9Vq1ZNY8eO1fXr1x/q2m3bttWuXbs0bNgwzZgxQxUrVnyo62Wmdu3ayWQyycfH557fxwMHDshkMslkMumTTz6x+/4nTpzQwIEDtWPHjgyIFgCQ2bIYHQAA3Mu3336rZs2aycPDQ23atNETTzyhGzduaMOGDXrnnXf0+++/6/PPP38oa1+/fl2bNm3S+++/r27duj2UNYKDg3X9+nW5ubk9lPv/kyxZsujatWtasmSJmjdvbnNt5syZ8vT0VEJCwgPd+8SJExo0aJAKFiyosmXLpvt9K1eufKD1AAAZiwQBgMM5fPiwWrZsqeDgYK1Zs0Z58uSxXOvatav++OMPffvttw9t/TNnzkiS/Pz8HtoaJpNJnp6eD+3+/8TDw0PVqlXT7NmzUyUIs2bNUv369bVgwYJMieXatWvKmjWr3N3dM2U9AMD9scUIgMMZMWKErly5oujoaJvk4I4iRYrozTfftHx969YtDRkyRI8//rg8PDxUsGBBvffee0pMTLR5X8GCBfXCCy9ow4YNevLJJ+Xp6anChQvrf//7n2XOwIEDFRwcLEl65513ZDKZVLBgQUkpW3Pu/NnawIEDZTKZbMZWrVql6tWry8/PT9mzZ1dISIjee+89y/W0ehDWrFmjGjVqKFu2bPLz81PDhg21d+/ee673xx9/qF27dvLz85Ovr6/at2+va9eupf2N/ZuXX35Z3333nS5evGgZ27Jliw4cOKCXX3451fzz58/r7bffVqlSpZQ9e3b5+PioXr16+u233yxz1q5dq0qVKkmS2rdvb9mqdOdz1qpVS0888YS2bt2qmjVrKmvWrJbvy997ENq2bStPT89Un79u3bry9/fXiRMn0v1ZAQDpR4IAwOEsWbJEhQsX1lNPPZWu+Z06dVL//v1Vvnx5jR49WmFhYYqKilLLli1Tzf3jjz/UtGlT1alTRyNHjpS/v7/atWun33//XZIUERGh0aNHS5JeeuklzZgxQ2PGjLEr/t9//10vvPCCEhMTNXjwYI0cOVIvvviifvrpp/u+7/vvv1fdunV1+vRpDRw4UJGRkdq4caOqVaumI0eOpJrfvHlz/fXXX4qKilLz5s0VExOjQYMGpTvOiIgImUwmLVy40DI2a9YshYaGqnz58qnmHzp0SIsXL9YLL7ygUaNG6Z133tGuXbsUFhZm+WG9ePHiGjx4sCSpS5cumjFjhmbMmKGaNWta7nPu3DnVq1dPZcuW1ZgxY1S7du17xjd27FjlzJlTbdu2VVJSkiTps88+08qVK/Xpp58qb9686f6sAAA7mAHAgVy6dMksydywYcN0zd+xY4dZkrlTp04242+//bZZknnNmjWWseDgYLMk8/r16y1jp0+fNnt4eJjfeusty9jhw4fNkswff/yxzT3btm1rDg4OThXDgAEDzNb/nI4ePdosyXzmzJk0476zxrRp0yxjZcuWNefKlct87tw5y9hvv/1mdnFxMbdp0ybVeh06dLC5Z+PGjc2BgYFprmn9ObJly2Y2m83mpk2bmp955hmz2Ww2JyUlmYOCgsyDBg265/cgISHBnJSUlOpzeHh4mAcPHmwZ27JlS6rPdkdYWJhZknny5Mn3vBYWFmYztmLFCrMk89ChQ82HDh0yZ8+e3dyoUaN//IwAgAdHBQGAQ7l8+bIkydvbO13zly1bJkmKjIy0GX/rrbckKVWvQokSJVSjRg3L1zlz5lRISIgOHTr0wDH/3Z3eha+//lrJycnpes/Jkye1Y8cOtWvXTgEBAZbx0qVLq06dOpbPae21116z+bpGjRo6d+6c5XuYHi+//LLWrl2r+Ph4rVmzRvHx8ffcXiSl9C24uKT830ZSUpLOnTtn2T61bdu2dK/p4eGh9u3bp2vus88+q1dffVWDBw9WRESEPD099dlnn6V7LQCA/UgQADgUHx8fSdJff/2Vrvl//vmnXFxcVKRIEZvxoKAg+fn56c8//7QZL1CgQKp7+Pv768KFCw8YcWotWrRQtWrV1KlTJ+XOnVstW7bU3Llz75ss3IkzJCQk1bXixYvr7Nmzunr1qs343z+Lv7+/JNn1WZ5//nl5e3vrq6++0syZM1WpUqVU38s7kpOTNXr0aBUtWlQeHh7KkSOHcubMqZ07d+rSpUvpXjNfvnx2NSR/8sknCggI0I4dOzRu3DjlypUr3e8FANiPBAGAQ/Hx8VHevHm1e/duu9739ybhtLi6ut5z3Gw2P/Aad/bH3+Hl5aX169fr+++/1yuvvKKdO3eqRYsWqlOnTqq5/8a/+Sx3eHh4KCIiQtOnT9eiRYvSrB5I0vDhwxUZGamaNWvqyy+/1IoVK7Rq1SqVLFky3ZUSKeX7Y4/t27fr9OnTkqRdu3bZ9V4AgP1IEAA4nBdeeEEHDx7Upk2b/nFucHCwkpOTdeDAAZvxU6dO6eLFi5YnEmUEf39/myf+3PH3KoUkubi46JlnntGoUaO0Z88eDRs2TGvWrNEPP/xwz3vfiTM2NjbVtX379ilHjhzKli3bv/sAaXj55Ze1fft2/fXXX/ds7L5j/vz5ql27tqKjo9WyZUs9++yzCg8PT/U9SW+ylh5Xr15V+/btVaJECXXp0kUjRozQli1bMuz+AIDUSBAAOJx3331X2bJlU6dOnXTq1KlU1w8ePKixY8dKStkiIynVk4ZGjRolSapfv36GxfX444/r0qVL2rlzp2Xs5MmTWrRokc288+fPp3rvnQPD/v7o1Tvy5MmjsmXLavr06TY/cO/evVsrV660fM6HoXbt2hoyZIjGjx+voKCgNOe5urqmqk7MmzdPx48ftxm7k8jcK5myV+/evRUXF6fp06dr1KhRKliwoNq2bZvm9xEA8O9xUBoAh/P4449r1qxZatGihYoXL25zkvLGjRs1b948tWvXTpJUpkwZtW3bVp9//rkuXryosLAw/fLLL5o+fboaNWqU5iM0H0TLli3Vu3dvNW7cWD169NC1a9c0adIkFStWzKZJd/DgwVq/fr3q16+v4OBgnT59WhMnTtRjjz2m6tWrp3n/jz/+WPXq1VPVqlXVsWNHXb9+XZ9++ql8fX01cODADPscf+fi4qJ+/fr947wXXnhBgwcPVvv27fXUU09p165dmjlzpgoXLmwz7/HHH5efn58mT54sb29vZcuWTZUrV1ahQoXsimvNmjWaOHGiBgwYYHns6rRp01SrVi198MEHGjFihF33AwCkDxUEAA7pxRdf1M6dO9W0aVN9/fXX6tq1q/r06aMjR45o5MiRGjdunGXulClTNGjQIG3ZskU9e/bUmjVr1LdvX82ZMydDYwoMDNSiRYuUNWtWvfvuu5o+fbqioqLUoEGDVLEXKFBAU6dOVdeuXTVhwgTVrFlTa9aska+vb5r3Dw8P1/LlyxUYGKj+/fvrk08+UZUqVfTTTz/Z/cP1w/Dee+/prbfe0ooVK/Tmm29q27Zt+vbbb5U/f36beW5ubpo+fbpcXV312muv6aWXXtK6devsWuuvv/5Shw4dVK5cOb3//vuW8Ro1aujNN9/UyJEj9fPPP2fI5wIA2DKZ7elmAwAAAPCfRgUBAAAAgAUJAgAAAAALEgQAAAAAFiQIAAAAACxIEAAAAABYkCAAAAAAsCBBAAAAAGDxnzxJOWuTqUaHADySDk5tbXQIwCMp8Vay0SEAj5yCgZ5Gh5Amr3LdMm2t69vHZ9pa6UUFAQAAAIDFf7KCAAAAADwwk3P/Dt25Pz0AAAAAG1QQAAAAAGsmk9ERGIoKAgAAAAALKggAAACANXoQAAAAACAFFQQAAADAGj0IAAAAAJCCCgIAAABgjR4EAAAAAEhBBQEAAACwRg8CAAAAAKSgggAAAABYowcBAAAAAFKQIAAAAACwYIsRAAAAYI0mZQAAAABIQQUBAAAAsEaTMgAAAACkoIIAAAAAWKMHAQAAAABSUEEAAAAArNGDAAAAAAApqCAAAAAA1uhBAAAAAIAUVBAAAAAAa/QgAAAAAEAKKggAAACANSoIAAAAAJCCCgIAAABgzYWnGAEAAACAJCoIAAAAgC16EAAAAAAgBQkCAAAAAAu2GAEAAADWTDQpAwAAAIAkKggAAACALZqUAQAAACAFFQQAAADAGj0IAAAAAJCCCgIAAABgjR4EAAAAAEhBBQEAAACwRg8CAAAAAKSgggAAAABYowcBAAAAAFJQQQAAAACs0YMAAAAAACmoIAAAAADW6EEAAAAAgBRUEAAAAABr9CAAAAAAQAoqCAAAAIA1ehAAAAAAIAUJAgAAAAALthgBAAAA1thiBAAAAAApSBAAAAAAayZT5r3sdPz4cbVu3VqBgYHy8vJSqVKl9Ouvv1qum81m9e/fX3ny5JGXl5fCw8N14MABu9YwPEG4efNmmtfOnj2biZEAAAAAjuvChQuqVq2a3Nzc9N1332nPnj0aOXKk/P39LXNGjBihcePGafLkydq8ebOyZcumunXrKiEhId3rGN6D0LJlS82fP1+mv2VQp06d0jPPPKPdu3cbFBkAAACckoP2IHz00UfKnz+/pk2bZhkrVKiQ5c9ms1ljxoxRv3791LBhQ0nS//73P+XOnVuLFy9Wy5Yt07WO4Z8+Li5OnTp1shmLj49XrVq1FBoaalBUAAAAwMOXmJioy5cv27wSExPvOfebb75RxYoV1axZM+XKlUvlypXTF198Ybl++PBhxcfHKzw83DLm6+urypUra9OmTemOyfAEYdmyZdq4caMiIyMlSSdOnFBYWJhKlSqluXPnGhwdAAAAnE4m9iBERUXJ19fX5hUVFXXPsA4dOqRJkyapaNGiWrFihV5//XX16NFD06dPl5TyS3ZJyp07t837cufObbmWHoZvMcqZM6dWrlyp6tWrS5KWLl2q8uXLa+bMmXJxMTx/AQAAAB6avn37Wn5RfoeHh8c95yYnJ6tixYoaPny4JKlcuXLavXu3Jk+erLZt22ZYTA7xE3j+/Pm1atUqzZw5U08++aRmz54tV1dXo8MCAACAMzK5ZNrLw8NDPj4+Nq+0EoQ8efKoRIkSNmPFixdXXFycJCkoKEhSSi+vtVOnTlmupYchFQR/f/9UTcmSdO3aNS1ZskSBgYGWsfPnz2dmaAAAAIBDqlatmmJjY23G9u/fr+DgYEkpDctBQUFavXq1ypYtK0m6fPmyNm/erNdffz3d6xiSIIwZM8aIZQEAAIB/9gDnE2SGXr166amnntLw4cPVvHlz/fLLL/r888/1+eefS5JMJpN69uypoUOHqmjRoipUqJA++OAD5c2bV40aNUr3OoYkCBm5RwoAAABwBpUqVdKiRYvUt29fDR48WIUKFdKYMWPUqlUry5x3331XV69eVZcuXXTx4kVVr15dy5cvl6enZ7rXMZnNZvPD+ADptW3bNrm5ualUqVKSpK+//lrTpk1TiRIlNHDgQLm7u9t9z6xNpmZ0mIBTODi1tdEhAI+kxFvJRocAPHIKBqb/B9bMlpk/S15b0CHT1kovw5uUX331Ve3fv19SyqObWrRooaxZs2revHl69913DY4OAAAAcC6GJwj79++3NFHMmzdPYWFhmjVrlmJiYrRgwQJjgwMAAIDTMZlMmfZyRIYnCGazWcnJKaXZ77//Xs8//7yklEefnj171sjQAAAAAKdj+EFpFStW1NChQxUeHq5169Zp0qRJklKOiv77KXAAAADAQ+eYv9jPNIZXEMaMGaNt27apW7duev/991WkSBFJ0vz58/XUU08ZHB0AAADgXAyvIJQuXVq7du1KNf7xxx9zmjIAAACQyQxPENJiz7NaAQAAgIziqM3DmcWQBCEgIED79+9Xjhw55O/vf9//Es6fP5+JkQEAAADOzZAEYfTo0fL29paU0oMAAAAAOAoqCAZo27btPf8MAAAAwFiGP8UoLdu2bdMLL7xgdBgAAABwMhyUZqAVK1bo7bff1nvvvadDhw5Jkvbt26dGjRqpUqVKlgPUAAAAAGQOw55iFB0drc6dOysgIEAXLlzQlClTNGrUKHXv3l0tWrTQ7t27Vbx4caPCAwAAgJNy1N/sZxbDKghjx47VRx99pLNnz2ru3Lk6e/asJk6cqF27dmny5MkkBwAAAIABDKsgHDx4UM2aNZMkRUREKEuWLPr444/12GOPGRUSMsj7zcvp/RblbMZij19UuR4LJUmFcnsrqu2TqhqaSx5urlq147jemrJJpy8lGBEu4NDOnD6lz8aP1i8bNyghMUH5Hsuv3h8MVWiJkkaHBjispKQkfRk9SatXfKsL584pMEdO1an/ol5u18XpfzOMdHLy/5kYliBcv35dWbNmlZRSxvHw8FCePHmMCgcZ7Pe4C3ph0HLL17eSUvpJsnpk0ZL+dbXryHk9PzDlev+Xymt+3zoK67tEZrMh4QIO6a/Ll9StcxuVq1BJH42dJD8/fx07GidvHx+jQwMc2twvp2nponl6u98QBRd+XAf27tHI4f2VLVt2NWreyujwAIdn6EnKU6ZMUfbs2SVJt27dUkxMjHLkyGEzp0ePHkaEhn8pKSlZpy5eTzVeNTSXgnNmV9W3v9Zf129Kkjp/ul4nprdWrVJ59cPOE5kdKuCwZv1vqnLlClKf/kMtY3nyUWUF/smeXTtUtUYtVa5WU5IUlCeffvj+O8Xu2W1wZHhUOHulybAEoUCBAvriiy8sXwcFBWnGjBk2c0wmEwnCI+rxPD46+EVLJdxM0ubY0+o/81cdO3tVHm6uMktKvJlkmZtwI0nJZrOeCs1NggBY2fjjWlWq/JQG9InUb9u3KkfOXGrUtIVeaNTU6NAAh1aiVFl99/UCHYs7oscKFNTBA7H6/bfterXH20aHBjwSDEsQjhw5YtTSeMi2HDijLuN/1IETlxTkn1XvNSur74fWV8WeC/XL/jO6mnBLQ1+ppAEzf5XJZNKQ1hWVxdVFQf5eRocOOJQTx4/p64Vz1fzlNmrdvrP27dmtcSM/VJYsbnruhYZGhwc4rBavdNC1q1fU6aVGcnFxVXJyktq92l1P161vdGh4RFBBeMQlJiYqMTHRZsycdFMmVzeDIsLK7ccsf9795wVt2X9G+yY3V5NqhTR99QG1HrlGY7s8pTeeL6Fks1lzNxzS9oNnlUwDAmDDnJyskOIl1fmNNyVJRUOK6/DBP/TNwrkkCMB9rF+9QmtWLlOfgVEKLlxEB/fv0+SxH6c0Kz//otHhAQ7vkU8QoqKiNGjQIJuxLKEvyq0E/+fpKC5du6E/Tl5S4aCUxsrVv53QE13nK9DbQ7eSzLp07YYOT2mpw6f+MjhSwLEE5sip4EKP24wFFyys9T98b1BEwKPhiwmj1eKVDqpVp54kqdDjRXU6/qTm/C+aBAHp4uwVBENPUs4Iffv21aVLl2xeWUKeNzosWMnmmUWFcvso/oJt0/K5vxJ16doNhT2RRzl9vfTtljiDIgQc0xOly+ron0dsxo7GHVHuIJ74BtxPYkKCTCbbH3FcXF1lNicbFBHwaHnkKwgeHh7y8PCwGWN7kbGGt6mkZb8eVdyZK8oTkFX9WpRTUnKy5m04JEl6pXZR7Tt2UWcvJ6hySC593KGyPl36uw6cuGxw5IBjafZyG3Xt+Iq+nPaFaoXX1b7fd2np4gV6673+RocGOLQq1cM0Z/oXypU7SMGFH9fB/fu0cM4MPVuf3QVIH2evIDzyCQIcT77AbJreq5YCvD109nKCNu49pVp9l+rs5ZSD0Irm89XgVhXkn91Df565ohELftOnS343NmjAAYWWeEJDRozRFxPHaHr0ZOXJm0/dIt9VnedeMDo0wKG90auPpn8xQeM/Ga6LF84rMEdOPd+wqVp1eNXo0IBHgslsNrYztE2bNqpdu7Zq1qypxx9//J/fkA5Zm0zNkPsAzubg1NZGhwA8khJvsXUFsFfBQE+jQ0hTYNvZmbbWuekvZdpa6WV4D4K7u7uioqJUtGhR5c+fX61bt9aUKVN04MABo0MDAAAAnI7hCcKUKVO0f/9+HT16VCNGjFD27Nk1cuRIhYaG6rHHODEUAAAAyEwO04Pg7++vwMBA+fv7y8/PT1myZFHOnDmNDgsAAABOxtmblA2vILz33nt66qmnFBgYqD59+ighIUF9+vRRfHy8tm/fbnR4AAAAgFMxvILw4YcfKmfOnBowYIAiIiJUrFgxo0MCAACAE3P2CoLhCcL27du1bt06rV27ViNHjpS7u7vCwsJUq1Yt1apVi4QBAAAAyESGJwhlypRRmTJl1KNHD0nSb7/9ptGjR6tr165KTk5WUlKSwRECAADAmVBBMJjZbNb27du1du1arV27Vhs2bNDly5dVunRphYWFGR0eAAAA4FQMTxACAgJ05coVlSlTRmFhYercubNq1KghPz8/o0MDAACAM3LuAoLxCcKXX36pGjVqyMfHx+hQAAAAAKdneIJQv359y5+PHTsmSRyQBgAAAMM4ew+C4ecgJCcna/DgwfL19VVwcLCCg4Pl5+enIUOGKDk52ejwAAAAAKdieAXh/fffV3R0tD788ENVq1ZNkrRhwwYNHDhQCQkJGjZsmMERAgAAwJk4ewXB8ARh+vTpmjJlil588UXLWOnSpZUvXz698cYbJAgAAABAJjI8QTh//rxCQ0NTjYeGhur8+fMGRAQAAABn5uwVBMN7EMqUKaPx48enGh8/frzKlCljQEQAAACA8zK8gjBixAjVr19f33//vapWrSpJ2rRpk44ePaply5YZHB0AAACcDRUEg4WFhWn//v1q3LixLl68qIsXLyoiIkKxsbGqUaOG0eEBAAAATsXwCoIk5c2bN1Uz8rFjx9SlSxd9/vnnBkUFAAAAp+TcBQTjKwhpOXfunKKjo40OAwAAAHAqDpsgAAAAAMh8DrHFCAAAAHAUNCkDAAAAwG2GVRAiIiLue/3ixYuZEwgAAABgxdkrCIYlCL6+vv94vU2bNpkUDQAAAADJwARh2rRpRi0NAAAApMnZKwj0IAAAAACw4ClGAAAAgDXnLiBQQQAAAABwFxUEAAAAwAo9CAAAAABwGxUEAAAAwAoVBAAAAAC4jQoCAAAAYIUKAgAAAADcRgUBAAAAsEIFAQAAAABuo4IAAAAAWHPuAgIVBAAAAAB3UUEAAAAArNCDAAAAAAC3kSAAAAAAsGCLEQAAAGCFLUYAAAAAcBsJAgAAAGDFZMq8lz0GDhwok8lk8woNDbVcT0hIUNeuXRUYGKjs2bOrSZMmOnXqlN2fnwQBAAAAeESULFlSJ0+etLw2bNhgudarVy8tWbJE8+bN07p163TixAlFRETYvQY9CAAAAICVzOxBSExMVGJios2Yh4eHPDw87jk/S5YsCgoKSjV+6dIlRUdHa9asWXr66aclSdOmTVPx4sX1888/q0qVKumOiQoCAAAAYJCoqCj5+vravKKiotKcf+DAAeXNm1eFCxdWq1atFBcXJ0naunWrbt68qfDwcMvc0NBQFShQQJs2bbIrJioIAAAAgJXMfIhR3759FRkZaTOWVvWgcuXKiomJUUhIiE6ePKlBgwapRo0a2r17t+Lj4+Xu7i4/Pz+b9+TOnVvx8fF2xUSCAAAAABjkftuJ/q5evXqWP5cuXVqVK1dWcHCw5s6dKy8vrwyLiS1GAAAAgJW/PynoYb7+DT8/PxUrVkx//PGHgoKCdOPGDV28eNFmzqlTp+7Zs3A/JAgAAADAI+jKlSs6ePCg8uTJowoVKsjNzU2rV6+2XI+NjVVcXJyqVq1q133ZYgQAAABYcdSDlN9++201aNBAwcHBOnHihAYMGCBXV1e99NJL8vX1VceOHRUZGamAgAD5+Pioe/fuqlq1ql1PMJJIEAAAAIBHwrFjx/TSSy/p3Llzypkzp6pXr66ff/5ZOXPmlCSNHj1aLi4uatKkiRITE1W3bl1NnDjR7nVIEAAAAAArLi6OWUKYM2fOfa97enpqwoQJmjBhwr9ahx4EAAAAABZUEAAAAAArjtqDkFmoIAAAAACwoIIAAAAAWPm35xM86qggAAAAALAgQQAAAABgwRYjAAAAwIqT7zCiggAAAADgLioIAAAAgBWalAEAAADgNioIAAAAgBUqCAAAAABwGxUEAAAAwIqTFxCoIAAAAAC4iwoCAAAAYIUeBAAAAAC4jQoCAAAAYMXJCwhUEAAAAADcRQUBAAAAsEIPAgAAAADcRgUBAAAAsOLkBQQqCAAAAADuooIAAAAAWKEHAQAAAABuo4IAAAAAWHHyAgIVBAAAAAB3kSAAAAAAsGCLEQAAAGCFJmUAAAAAuO0/WUE4O7u90SEAj6TAmn2MDgF4JC2e+rbRIQCPnIKBnkaHkCYnLyBQQQAAAABw13+yggAAAAA8KHoQAAAAAOA2KggAAACAFScvIFBBAAAAAHAXFQQAAADACj0IAAAAAHAbFQQAAADAipMXEKggAAAAALiLCgIAAABghR4EAAAAALiNCgIAAABghQoCAAAAANxGBQEAAACw4uQFBCoIAAAAAO4iQQAAAABgwRYjAAAAwApNygAAAABwGxUEAAAAwIqTFxCoIAAAAAC4iwoCAAAAYIUeBAAAAAC4jQoCAAAAYMXJCwhUEAAAAADcRQUBAAAAsOLi5CUEKggAAAAALKggAAAAAFacvIBABQEAAADAXVQQAAAAACucgwAAAAAAt1FBAAAAAKy4OHcBgQoCAAAAgLuoIAAAAABW6EEAAAAAgNuoIAAAAABWnLyAQAUBAAAAwF0kCAAAAAAs2GIEAAAAWDHJufcYUUEAAAAAHjEffvihTCaTevbsaRlLSEhQ165dFRgYqOzZs6tJkyY6deqU3fcmQQAAAACsuJgy7/UgtmzZos8++0ylS5e2Ge/Vq5eWLFmiefPmad26dTpx4oQiIiLs//wPFhYAAACAzHblyhW1atVKX3zxhfz9/S3jly5dUnR0tEaNGqWnn35aFSpU0LRp07Rx40b9/PPPdq1BggAAAABYMZlMmfZKTEzU5cuXbV6JiYlpxta1a1fVr19f4eHhNuNbt27VzZs3bcZDQ0NVoEABbdq0ya7PT4IAAAAAGCQqKkq+vr42r6ioqHvOnTNnjrZt23bP6/Hx8XJ3d5efn5/NeO7cuRUfH29XTDzFCAAAALCSmQel9e3bV5GRkTZjHh4eqeYdPXpUb775platWiVPT8+HGhMJAgAAAGAQDw+PeyYEf7d161adPn1a5cuXt4wlJSVp/fr1Gj9+vFasWKEbN27o4sWLNlWEU6dOKSgoyK6YSBAAAAAAKy6ZWUJIp2eeeUa7du2yGWvfvr1CQ0PVu3dv5c+fX25ublq9erWaNGkiSYqNjVVcXJyqVq1q11okCAAAAICD8/b21hNPPGEzli1bNgUGBlrGO3bsqMjISAUEBMjHx0fdu3dX1apVVaVKFbvWIkEAAAAArDhgASFdRo8eLRcXFzVp0kSJiYmqW7euJk6caPd9SBAAAACAR9DatWttvvb09NSECRM0YcKEf3VfEgQAAADAiulRLSFkEM5BAAAAAGBBBQEAAACw4uQFBCoIAAAAAO6iggAAAABYccRzEDITFQQAAAAAFiQIAAAAACzYYgQAAABYce4NRlQQAAAAAFihggAAAABY4aA0AAAAALiNCgIAAABgxcW5CwhUEAAAAADcZXiCsHv37jSvLV68OPMCAQAAAJTSg5BZL0dkeIJQt25dHT58ONX4ggUL1KpVKwMiAgAAAJyX4QlCp06dFB4ervj4eMvYV199pTZt2igmJsa4wAAAAOCUTKbMezkiw5uUBw0apPPnzys8PFzr16/X8uXL1alTJ82YMUNNmjQxOjwAAADAqRieIEjSp59+qlatWqlKlSo6fvy4Zs+erYYNGxodFgAAAJyQo/YGZBZDEoRvvvkm1VhERIR+/PFHvfTSSzKZTJY5L774YmaHBwAAADgtQxKERo0apXlt6tSpmjp1qqSU7C0pKSmTogIAAAA4B8GQBCE5OdmIZQEAAAD8A4foQfi7ixcvys/Pz+gwAAAA4IToQUiHe/UMpMXenoGPPvpIBQsWVIsWLSRJzZo104IFC5QnTx4tW7ZMZcqUset+AAAAAB5cuhKE+/UMWHuQnoHJkydr5syZkqRVq1bp+++/1/LlyzV37ly98847WrlypV33AwAAAP4N564fpDNBeJg9A/Hx8cqfP78kaenSpWrevLmeffZZFSxYUJUrV35o6wIAAABIzfCTlP39/XX06FFJ0vLlyxUeHi5JMpvNPMEIAAAAmc7FZMq0lyN6oCblq1evat26dYqLi9ONGzdsrvXo0cOue0VEROjll19W0aJFde7cOdWrV0+StH37dhUpUuRBwgMAAADwgOxOELZv367nn39e165d09WrVxUQEKCzZ88qa9asypUrl90JwujRo1WwYEEdPXpUI0aMUPbs2SVJJ0+e1BtvvGFveAAAAAD+BbsThF69eqlBgwaaPHmyfH199fPPP8vNzU2tW7fWm2++aXcAbm5uevvtt++5DgAAAJDZHHTnT6axO0HYsWOHPvvsM7m4uMjV1VWJiYkqXLiwRowYobZt2yoiIuIf7/HNN9+oXr16cnNz+8dHqNr72FQAAAAAD87uBMHNzU0uLim9zbly5VJcXJyKFy8uX19fS7PxP2nUqJHi4+OVK1eu+z5C9UEemwoAAAD8GxyUZqdy5cppy5YtKlq0qMLCwtS/f3+dPXtWM2bM0BNPPJGue1g/NvVhPkIVAAAAgH3sfszp8OHDlSdPHknSsGHD5O/vr9dff11nzpzR559/nqHBHT9+PEPvBwAAAPwTkynzXo7I7gpCxYoVLX/OlSuXli9fnqEBSSmHpw0bNkzR0dG6du1aht8fAAAAwL0ZdlDahQsX9NJLLylHjhzKmzevxo0bp+TkZPXv31+FCxfWli1bNG3aNKPCAwAAgJPioDQ7FSpU6L6NG4cOHUrXffr06aONGzeqXbt2WrFihXr16qXly5fLxcVFa9asUZUqVewNDQ5s669b9L+YaO3Z87vOnjmjUWPGq/Yz4UaHBTiUvDl9NLRrPT1bNURZPdx18NhZvTp0nrbtu7vdMqRgLg3tWk81yhVWFlcX7Tt8Si/1/VJHT100LnDAQCsXzNDOn9fp1LE/5ebuoUKhpfRim9eVO18By5ybNxK1aNp4bduwWrdu3VTxsk+q2atvyccvwMDIAcdld4LQs2dPm69v3ryp7du3a/ny5XrnnXfSfZ/vvvtOMTExevrpp9WtWzcVLlxYZcuW1fDhw+0NCY+A69evq1ixUDVs3ERv9exudDiAw/Hz9tKaz1/Xuq2H1KjXVJ25cFVF8ufQhb+uW+YUyheg1Z+9pulLtmjoF6t0+WqCShTOrYQbNw2MHDDWH79vV416ESpQJFTJSUlaMvNzTRzUS++N+1Ienl6SpIVTP9WerRvV4Z0h8syWTfM/H63oj95Xr6hJBkcPR+Wgv9jPNHYnCGkdhjZhwgT9+uuv6b7PiRMnVLx4cUlSwYIF5enpqdatW9sbDh4R1WvUVPUaNY0OA3BYb70SpmOnLunVofMsY3+evGAzZ9Brz2nFxli9P/47y9jh4+czLUbAEb3Rf5TN1626v6f32zXQ0YOxKlKyrK5fvaKfVy9Vm14DVKx0BcucYd1b6XDsbhUKSd8TGAFnkmE9CPXq1dOCBQvSPd9sNitLlrv5iaurq7y8vDIqHAB4pNSvUULb9h7TzGGt9OeyD7Rpeg+1b/ik5brJZNJzT4XqQNxZfTOmo/5c9oHWR3dVg5olDIwacDwJ165KkrJm95EkHT0Yq6RbtxRS5u5DVnI/Fiz/nLl1JPZ3Q2KE4zOZTJn2ckR2VxDSMn/+fAUEpH8vn9ls1jPPPGNJEq5fv64GDRrI3d3dZt62bdvue5/ExEQlJibajCWZ3OXh4ZHuWADAaIXyBqhzRBWNm/2jRkz/QRWKP6aRvV7UjZu3NHPZNuXyzybvbB56u00tDfpshfpNWKZnq4RozoevqG7Xz7Vh+2GjPwJguOTkZC2MHqfCoaWUN7iwJOnyxXNyzeKmrNm8beZ6+wbo8sVzRoQJOLwHOijNOtsxm82Kj4/XmTNnNHHixHTfZ8CAATZfN2zY0N5QJElRUVEaNGiQzdh7/frr/Q8GPtD9AMAILi4mbdt7XAMmr5Ak/bb/hEo+HqTOjato5rJtcnFJ+Xd36frf9emcDZKknQdOqnLpYHVuXIUEAZA07/NROhl3SG8OT//PI8C9GPaYTwdhd4LQsGFDmwTBxcVFOXPmVK1atRQaGpru+/w9QXhQffv2VWRkpM1Yksk9jdkA4Jjiz/6lvUdO2YztO3JajWql7I8+e/Gabt5K0t4jp23mxB45rafKFMysMAGHNe/zUfr91416c9h4+efIZRn38QtU0q2bunb1L5sqwl+XzsvHL9CIUAGHZ3eCMHDgwIcQxoPz8PBItZ3o2g2zQdEAwIPZtPOIihXIaTNWNH8OxcVflCTdvJWkrXuO3XvO35qZAWdiNps1/4vR2rl5vboP+VSBufPaXM//eIhcs2TR/p1bVbZqLUnSqeNxunDmlAqGlDQgYjwKHLU3ILPYXUFxdXXV6dOnU42fO3dOrq6uGRIU/nuuXbuq2H17FbtvryTp+PFjit23VydPnjA4MsAxfDpng558ooDeaVtbhR8LVItny6pDo8r6bMFGy5zRM9epaXhptW/4pAo/FqjXmlbV89WL6/OFPxsYOWCseZ+P1K/rVqpNrwHy9MqqyxfO6fKFc7pxuz/RK1t2VXnmBS2a9qn279qmuIP7NPPT4SoY8gRPMALSYDKbzXb9ut3FxUXx8fHKlSuXzfiJEyf0+OOP6/r162m8M/NQQXA8v27ZrM4d2qYab/BiIw0e9qEBEeFeAmv2MToEp1avWqgGv/6ciuTPoSMnL2jc7B817etfbOa0eaGi3mlbW/ly+mp/3BkN/WKVlv64x6CIccfiqW8bHYLT6tG4+j3HW3V/T5Wffl6S9UFp3+vWzZsKLfukmr/6lnz82WJkpLolcv7zJIP0/Hpfpq01pmH6t+hnlnQnCOPGjZMk9erVS0OGDFH27Nkt15KSkrR+/XodOXJE27dvfziR2oEEAXgwJAjAgyFBAOxHgpDCEROEdPcgjB49WlLKXr/JkyfbbCdyd3dXwYIFNXny5IyPEAAAAECmSXeCcPhwyiP0ateurYULF8rf3/+BF71TjUiPHj16PPA6AAAAgL1cnLtH2f6nGP3www//etE71Yg7zpw5o2vXrsnPz0+SdPHiRWXNmlW5cuUiQQAAAAAykd1PMWrSpIk++uijVOMjRoxQs2bN0nWPw4cPW17Dhg1T2bJltXfvXp0/f17nz5/X3r17Vb58eQ0ZMsTe8AAAAIB/xWQyZdrLEdmdIKxfv17PP/98qvF69epp/fr1dgfwwQcf6NNPP1VISIhlLCQkRKNHj1a/fv3svh8AAACAB2f3FqMrV67I3T31ScVubm66fPmy3QGcPHlSt27dSjWelJSkU6dO3eMdAAAAwMPj7D0IdlcQSpUqpa+++irV+Jw5c1SiRAm7A3jmmWf06quvatu2bZaxrVu36vXXX1d4eLjd9wMAAADw4OyuIHzwwQeKiIjQwYMH9fTTT0uSVq9erVmzZmn+/Pl2BzB16lS1bdtWFStWlJubmyTp1q1bqlu3rqZMmWL3/QAAAIB/w0FbAzKN3QlCgwYNtHjxYg0fPlzz58+Xl5eXypQpozVr1iggIMDuAHLmzKlly5Zp//792rcv5VCK0NBQFStWzO57AQAAAPh37E4QJKl+/fqqX7++JOny5cuaPXu23n77bW3dulVJSUkPFEixYsVICgAAAGA4FycvITxQgiClPM0oOjpaCxYsUN68eRUREaEJEyak672RkZEaMmSIsmXLpsjIyPvOHTVq1IOGCAAAAMBOdiUI8fHxiomJUXR0tC5fvqzmzZsrMTFRixcvtqtBefv27bp586blz2lx1GfDAgAA4L/L7qf4/MekO0Fo0KCB1q9fr/r162vMmDF67rnn5OrqqsmTJ9u9qPVpzBlxMjMAAACAjJHuBOG7775Tjx499Prrr6to0aIPMyYAAADAMM6+iSXdCcKGDRsUHR2tChUqqHjx4nrllVfUsmXLB164Q4cO6Zo3derUB14DAAAAgH3SnSBUqVJFVapU0ZgxY/TVV19p6tSpioyMVHJyslatWqX8+fPL29s73QvHxMQoODhY5cqVk9lsfqDgAQAAgIzGU4zslC1bNnXo0EEdOnRQbGysoqOj9eGHH6pPnz6qU6eOvvnmm3Td5/XXX9fs2bN1+PBhtW/fXq1bt36gcxQAAAAAZJx/1aQdEhKiESNG6NixY5o9e7Zd750wYYJOnjypd999V0uWLFH+/PnVvHlzrVixgooCAAAADGMyZd7LEWXIU5xcXV3VqFGjdFcP7vDw8NBLL72kVatWac+ePSpZsqTeeOMNFSxYUFeuXMmI0AAAAADY4YEPSstoLi4uMplMMpvND3waMwAAAPBvuTjob/Yzi6HnQCQmJmr27NmqU6eOihUrpl27dmn8+PGKi4tT9uzZjQwNAAAAcEqGVRDeeOMNzZkzR/nz51eHDh00e/Zs5ciRw6hwAAAAAMjABGHy5MkqUKCAChcurHXr1mndunX3nLdw4cJMjgwAAADOzFEfczpp0iRNmjRJR44ckSSVLFlS/fv3V7169SRJCQkJeuuttzRnzhwlJiaqbt26mjhxonLnzm3XOoYlCG3atJHJQb/5AAAAgKN57LHH9OGHH6po0aIym82aPn26GjZsqO3bt6tkyZLq1auXvv32W82bN0++vr7q1q2bIiIi9NNPP9m1jsn8H3ym6LUb/7mPBGSKwJp9jA4BeCQtnvq20SEAj5y6JXIaHUKahnz/R6at9UF4kX/1/oCAAH388cdq2rSpcubMqVmzZqlp06aSpH379ql48eLatGmTqlSpku57GtqkDAAAADizxMREXb582eaVmJj4j+9LSkrSnDlzdPXqVVWtWlVbt27VzZs3FR4ebpkTGhqqAgUKaNOmTXbFRIIAAAAAWHExZd4rKipKvr6+Nq+oqKg0Y9u1a5eyZ88uDw8Pvfbaa1q0aJFKlCih+Ph4ubu7y8/Pz2Z+7ty5FR8fb9fnd5hzEAAAAABn07dvX0VGRtqMeXh4pDk/JCREO3bs0KVLlzR//ny1bds2zYf9PCgSBAAAAMCKSZn3IB0PD4/7JgR/5+7uriJFUvoWKlSooC1btmjs2LFq0aKFbty4oYsXL9pUEU6dOqWgoCC7YmKLEQAAAPCISk5OVmJioipUqCA3NzetXr3aci02NlZxcXGqWrWqXfekggAAAABYcXHQJ/H37dtX9erVU4ECBfTXX39p1qxZWrt2rVasWCFfX1917NhRkZGRCggIkI+Pj7p3766qVava9QQjiQQBAAAAeCScPn1abdq00cmTJ+Xr66vSpUtrxYoVqlOnjiRp9OjRcnFxUZMmTWwOSrMXCQIAAABgxVErCNHR0fe97unpqQkTJmjChAn/ah16EAAAAABYUEEAAAAArJhMDlpCyCRUEAAAAABYUEEAAAAArDhqD0JmoYIAAAAAwIIKAgAAAGDFyVsQqCAAAAAAuIsEAQAAAIAFW4wAAAAAKy5OvseICgIAAAAACyoIAAAAgBUecwoAAAAAt1FBAAAAAKw4eQsCFQQAAAAAd1FBAAAAAKy4yLlLCFQQAAAAAFhQQQAAAACs0IMAAAAAALdRQQAAAACscA4CAAAAANxGBQEAAACw4uLkTQhUEAAAAABYUEEAAAAArDh5AYEKAgAAAIC7qCAAAAAAVuhBAAAAAIDbqCAAAAAAVpy8gEAFAQAAAMBdJAgAAAAALNhiBAAAAFhx9t+gO/vnBwAAAGCFCgIAAABgxeTkXcpUEAAAAABYUEEAAAAArDh3/YAKAgAAAAArVBAAAAAAKy70IAAAAABACioIAAAAgBXnrh9QQQAAAABghQoCAAAAYMXJWxCoIAAAAAC4iwoCAAAAYIWTlAEAAADgNioIAAAAgBVn/w26s39+AAAAAFaoIAAAAABW6EEAAAAAgNtIEAAAAABYsMUIAAAAsOLcG4yoIAAAAACwQgUBAAAAsOLsTcr/yQTBxcW5/0sFHtS6We8ZHQLwSApr8r7RIQCPnOvbxxsdAtLwn0wQAAAAgAfl7Hvwnf3zAwAAALBCBQEAAACw4uw9CFQQAAAAAFhQQQAAAACsOHf9gAoCAAAAACtUEAAAAAArTt6CQAUBAAAAwF1UEAAAAAArLk7ehUAFAQAAAIAFFQQAAADACj0IAAAAAHAbFQQAAADAiokeBAAAAABIQQUBAAAAsEIPAgAAAADcRoIAAAAAPAKioqJUqVIleXt7K1euXGrUqJFiY2Nt5iQkJKhr164KDAxU9uzZ1aRJE506dcqudUgQAAAAACsuMmXayx7r1q1T165d9fPPP2vVqlW6efOmnn32WV29etUyp1evXlqyZInmzZundevW6cSJE4qIiLBrHXoQAAAAgEfA8uXLbb6OiYlRrly5tHXrVtWsWVOXLl1SdHS0Zs2apaefflqSNG3aNBUvXlw///yzqlSpkq51SBAAAAAAK5nZpJyYmKjExESbMQ8PD3l4ePzjey9duiRJCggIkCRt3bpVN2/eVHh4uGVOaGioChQooE2bNqU7QWCLEQAAAGCQqKgo+fr62ryioqL+8X3Jycnq2bOnqlWrpieeeEKSFB8fL3d3d/n5+dnMzZ07t+Lj49MdExUEAAAAwEpmVhD69u2ryMhIm7H0VA+6du2q3bt3a8OGDRkeEwkCAAAAYJD0biey1q1bNy1dulTr16/XY489ZhkPCgrSjRs3dPHiRZsqwqlTpxQUFJTu+7PFCAAAALBiysT/2MNsNqtbt25atGiR1qxZo0KFCtlcr1Chgtzc3LR69WrLWGxsrOLi4lS1atV0r0MFAQAAAHgEdO3aVbNmzdLXX38tb29vS1+Br6+vvLy85Ovrq44dOyoyMlIBAQHy8fFR9+7dVbVq1XQ3KEskCAAAAIANl0zsQbDHpEmTJEm1atWyGZ82bZratWsnSRo9erRcXFzUpEkTJSYmqm7dupo4caJd65AgAAAAAI8As9n8j3M8PT01YcIETZgw4YHXIUEAAAAArNjbG/BfQ5MyAAAAAAsqCAAAAICVzDwHwRFRQQAAAABgQQUBAAAAsEIPAgAAAADcRgUBAAAAsOKo5yBkFioIAAAAACxIEAAAAABYsMUIAAAAsEKTMgAAAADcRgUBAAAAsMJBaQAAAABwGxUEAAAAwIqTFxCoIAAAAAC4iwoCAAAAYMXFyZsQqCAAAAAAsKCCAAAAAFhx7voBFQQAAAAAVqggAAAAANacvIRABQEAAACABRUEAAAAwIrJyUsIVBAAAAAAWFBBAAAAAKw4+TEIVBAAAAAA3EUFAQAAALDi5AUEKggAAAAA7qKCAAAAAFhz8hICFQQAAAAAFiQIAAAAACzYYgQAAABY4aA0g/3www9pXpswYUImRgIAAADA8AQhIiJCW7duTTU+duxY9e3b14CIAAAA4MxMpsx7OSLDE4SPP/5Y9erV0759+yxjI0eOVP/+/fXtt98aGBkAAADgfAzvQejUqZPOnz+v8PBwbdiwQV999ZWGDx+uZcuWqVq1akaHBwAAACfjoL/YzzSGJwiS9O677+rcuXOqWLGikpKStGLFClWpUsXosAAAAACnY0iCMG7cuFRj+fLlU9asWVWzZk398ssv+uWXXyRJPXr0yOzwAAAA4MycvIRgMpvN5sxetFChQumaZzKZdOjQIbvvn3DL7rcAkLQz7pLRIQCPpLAm7xsdAvDIub59vNEhpGnbn5czba3ywT6ZtlZ6GVJBOHz4sBHLAgAAAP+IcxAcTFJSknbs2KELFy4YHQoAAADgdAxPEHr27Kno6GhJKclBzZo1Vb58eeXPn19r1641NjgAAAA4Hc5BMNj8+fNVpkwZSdKSJUt05MgR7du3T7169dL777OnEwAAAMhMhicIZ8+eVVBQkCRp2bJlatasmYoVK6YOHTpo165dBkcHAAAAZ2PKxJcjMjxByJ07t/bs2aOkpCQtX75cderUkSRdu3ZNrq6uBkcHAAAAOBfDD0pr3769mjdvrjx58shkMik8PFyStHnzZoWGhhocHQAAAJyOo/5qP5MYniAMHDhQTzzxhI4ePapmzZrJw8NDkuTq6qo+ffoYHB0AAADgXAxPECSpadOmqcbatm1rQCQAAABwds5+DoIhCcK4cePUpUsXeXp6aty4cfed26NHj0yKCgAAAIDJbDabM3vRQoUK6ddff1VgYKAKFSqU5jyTyaRDhw7Zff+EW/8mOsB57Yy7ZHQIwCMprAmP5QbsdX37eKNDSNPOo1cyba3S+bNn2lrpZUgF4fDhw/f8MwAAAGA0Rz3ALLMY/pjT+/n111+NDgEAAABwKoYnCFeuXNH169dtxnbs2KEGDRqocuXKBkUFAAAAZ8VBaQY5evSoqlatKl9fX/n6+ioyMlLXrl1TmzZtVLlyZWXLlk0bN240KjwAAADAKRn2mNN33nlHCQkJGjt2rBYuXKixY8fqxx9/VOXKlXXw4EE99thjRoUGAAAAZ+aov9rPJIYlCOvXr9fChQtVpUoVNW/eXEFBQWrVqpV69uxpVEgAAACA0zMsQTh16pTlEae5cuVS1qxZVa9ePaPCAQAAACRxUJqhJym7uLjY/Nnd3d3AaPCwzZk1U9OnRevs2TMqFhKqPu99oFKlSxsdFuAw9u3apm/nf6kjf+zTxfNn9eYHI1TxqVo2c47HHdZXU8dr365tSkpKUr4ChdSj30fKkSvImKABB5A3p6+GvtlQz1Yrqayebjp49KxeHfiltu2JkyRl83LX0B4N1aB2aQX4ZtORE+c0cfY6TZm/weDIAcdkWIJgNptVrFgxmW4/aPbKlSsqV66cTdIgSefPnzciPGSw5d8t0ycjotRvwCCVKlVGM2dM1+uvdtTXS5crMDDQ6PAAh5CYkKAChYsq7NkGGju0d6rrp04c09C3O6tm3RcV0bqLvLJm0/G4Q3LjlytwYn7eXloTE6l1Ww6oUbeJOnPhiooUyKkLl69Z5nz0VhPVqlRM7d//n/48cU7hVYtrbN/mOnnmkr5dt8vA6OGonP0cBMMShGnTphm1NAwwY/o0RTRtrkaNm0iS+g0YpPXr12rxwgXq2LmLwdEBjqFMpadUptJTaV6fN32SylSqppc69rCM5c7LAx3g3N5qX0fH4i/o1YFfWsb+PHHOZk6VMoX05dLN+nHrAUnS1IU/qWOTaqpYMpgEAbgHwxKEtm3bGrU0MtnNGze0d8/v6tj5VcuYi4uLqlR5Sjt/225gZMCjIzk5Wb9t+Un1m76iEe9315GD+5UzKK8aNG+bahsS4Ezqh5XS9xv3auaIDqpeoahOnL6oz+f+qGmL7j4q/effDuuFsFL63+JNOnHmkmpWLKqiwbn07sgFBkYOR+bkBQTjD0rDf9+FixeUlJSUaitRYGCgzp49a1BUwKPl8sXzSrh+TUvmTlepilXVe9inqvhULY0b2lt7d24zOjzAMIXy5VDnZjX0R9wZvfjGBH0xb4NGvttUrRrcPWw18qN52nsoXgdXDtPlX8bqmwlvqOeHc/XTtoMGRg44LkOblDNCYmKiEhMTbcbMrh7y8PAwKCIAyHhms1mSVKFqTdVr/LIkKfjxYjqwZ6fWLFuo4qXLGxkeYBgXF5O27YnTgPFLJEm/xR5TySJ51Llpdc1cslmS9EbLMD1ZqqCavDlZcSfPq3r5IhrTJ6UH4YfNsUaGD0fl5CWER76CEBUVZTmN+c7r44+ijA4LVvz9/OXq6qpz52z3hJ47d045cuQwKCrg0eLt4ydXV1flLVDIZjxv/oI6dybeoKgA48Wfvay9h2z/Duw7HK/8Qf6SJE8PNw3q3kC9Ry7UsvW7tfvACU3+ar3mr9ymnq88Y0TIgMN75BOEvn376tKlSzavd3r3NTosWHFzd1fxEiW1+edNlrHk5GRt3rxJpcuUMzAy4NGRxc1NhYqVUPyxOJvx+ONxPOIUTm3TjkMqFpzLZqxogVyKO5nyFES3LK5yd8ui5NtVuDuSkpLl4uLkvyZGmkyZ+B9H9MgnCB4eHvLx8bF5sb3I8bzStr0Wzp+rbxYv0qGDBzV08EBdv35djRpHGB0a4DASrl/Tnwf368+D+yVJZ06d0J8H9+vs6ZTfjtZv0lo/r1+lH75brFMnjmrVN3O1ffMGPVO/qZFhA4b69Ms1erJUIb3T4VkVzp9DLZ6rqA5Nqumzr9ZLkv66mqD1vx7Q8J6NVKNCUQXnDVTrBpXV6oUn9c0PvxkcPeCYTGbz31LqTJaUlKSYmBitXr1ap0+fVnJyss31NWvW2H3PhFsZFR0y0uyZX1oOSgsJLa7e7/VT6dJljA4LVnbGXTI6BKe2d+dWDe/9eqrx6uH19epbAyRJ61Z8oyVzp+v82dPK81gBRbTuogpVwzI7VPxNWJP3jQ7BqdWr8YQGd39RRQrk1JHj5zTuyzU2TzHKHeitwd0bKrxqqPx9siru5HlNXbhR4760/2cMZJzr28cbHUKaYuOv/fOkDBISlDXT1kovwxOEbt26KSYmRvXr11eePHksB6fdMXr0aLvvSYIAPBgSBODBkCAA9iNBSOGICYLhTzGaM2eO5s6dq+eff97oUAAAAACnZ3iC4O7uriJFihgdBgAAACDJ6Z9yanyT8ltvvaWxY8fK4J1OAAAAgENbv369GjRooLx588pkMmnx4sU2181ms/r37688efLIy8tL4eHhOnDggN3rGFJBiIiwfXLNmjVr9N1336lkyZJyc3OzubZw4cLMDA0AAADOzkFLCFevXlWZMmXUoUOHVD9PS9KIESM0btw4TZ8+XYUKFdIHH3ygunXras+ePfL09Ez3OoYkCL6+vjZfN27c2IgwAAAAAEMlJiYqMTHRZszDw+Oej+2vV6+e6tWrd8/7mM1mjRkzRv369VPDhg0lSf/73/+UO3duLV68WC1btkx3TIYkCNOmTTNiWQAAAOAfZeYBZlFRURo0aJDN2IABAzRw4EC77nP48GHFx8crPDzcMubr66vKlStr06ZNjp8gAAAAAJD69u2ryMhIm7EHOfQ3Pj7lUM3cuXPbjOfOndtyLb0MTxDKlSuX6uwDSTKZTPL09FSRIkXUrl071a5d24DoAAAA4Gzu8aPpQ5PWdiIjGf4Uo+eee06HDh1StmzZVLt2bdWuXVvZs2fXwYMHValSJZ08eVLh4eH6+uuvjQ4VAAAAcEhBQUGSpFOnTtmMnzp1ynItvQyvIJw9e1ZvvfWWPvjgA5vxoUOH6s8//9TKlSs1YMAADRkyxNJwAQAAADwsDvoQo/sqVKiQgoKCtHr1apUtW1aSdPnyZW3evFmvv/66XfcyvIIwd+5cvfTSS6nGW7Zsqblz50qSXnrpJcXGxmZ2aAAAAIDDuHLlinbs2KEdO3ZISmlM3rFjh+Li4mQymdSzZ08NHTpU33zzjXbt2qU2bdoob968atSokV3rGF5B8PT01MaNG1Odprxx40bL81qTk5PtenYrAAAA8MActITw66+/2vTl3mlubtu2rWJiYvTuu+/q6tWr6tKliy5evKjq1atr+fLldv8cbXiC0L17d7322mvaunWrKlWqJEnasmWLpkyZovfee0+StGLFCkupBAAAAHBGtWrVktlsTvO6yWTS4MGDNXjw4H+1jsl8v1UyycyZMzV+/HjLNqKQkBB1795dL7/8siTp+vXrlqcapUfCrYcWKvCftjPuktEhAI+ksCbvGx0C8Mi5vn280SGk6dCZhExbq3BOx9slY3gFQZJatWqlVq1apXndy8srE6MBAAAAnJdDJAgAAACAo8jMcxAckSEJQkBAgPbv368cOXLI39//ngel3XH+/PlMjAwAAABwboYkCKNHj5a3t7ckacyYMUaEAAAAANyTkxcQjEkQ2rZte88/AwAAADCWYT0Ily9fTtc8Hx+fhxwJAAAAYMXJSwiGJQh+fn737T0wm80ymUxKSkrKxKgAAAAA52ZYgvDDDz9Y/mw2m/X8889rypQpypcvn1EhAQAAAE7PsAQhLCzM5mtXV1dVqVJFhQsXNigiAAAAQDI5+R4jF6MDAAAAAOA4OCgNAAAAsOLsB6U5VAXhfk3LAAAAAB4+wyoIERERNl8nJCTotddeU7Zs2WzGFy5cmJlhAQAAwMk5+6+sDUsQfH19bb5u3bq1QZEAAAAAuMOwBGHatGlGLQ0AAACkydl3vTtUDwIAAAAAY/EUIwAAAMCGc5cQqCAAAAAAsKCCAAAAAFihBwEAAAAAbqOCAAAAAFhx8gICFQQAAAAAd1FBAAAAAKzQgwAAAAAAt1FBAAAAAKyYnLwLgQoCAAAAAAsSBAAAAAAWbDECAAAArDn3DiMqCAAAAADuooIAAAAAWHHyAgIVBAAAAAB3UUEAAAAArHBQGgAAAADcRgUBAAAAsMJBaQAAAABwGxUEAAAAwJpzFxCoIAAAAAC4iwoCAAAAYMXJCwhUEAAAAADcRQUBAAAAsMI5CAAAAABwGxUEAAAAwArnIAAAAADAbVQQAAAAACv0IAAAAADAbSQIAAAAACxIEAAAAABYkCAAAAAAsKBJGQAAALBCkzIAAAAA3EYFAQAAALDCQWkAAAAAcBsVBAAAAMAKPQgAAAAAcBsVBAAAAMCKkxcQqCAAAAAAuIsKAgAAAGDNyUsIVBAAAAAAWFBBAAAAAKxwDgIAAAAA3EYFAQAAALDCOQgAAAAAcBsVBAAAAMCKkxcQqCAAAAAAuIsKAgAAAGDNyUsIVBAAAAAAWJAgAAAAALAgQQAAAACsmDLxPw9iwoQJKliwoDw9PVW5cmX98ssvGfr5SRAAAACAR8RXX32lyMhIDRgwQNu2bVOZMmVUt25dnT59OsPWIEEAAAAArJhMmfey16hRo9S5c2e1b99eJUqU0OTJk5U1a1ZNnTo1wz4/CQIAAABgkMTERF2+fNnmlZiYeM+5N27c0NatWxUeHm4Zc3FxUXh4uDZt2pRhMf0nH3Pq+Z/8VP8NiYmJioqKUt++feXh4WF0OPibJwv7Gh0C7oG/N47v+vbxRoeAe+DvDh5UZv4sOXBolAYNGmQzNmDAAA0cODDV3LNnzyopKUm5c+e2Gc+dO7f27duXYTGZzGazOcPuBvyDy5cvy9fXV5cuXZKPj4/R4QCPBP7eAA+Gvzt4FCQmJqaqGHh4eNwzqT1x4oTy5cunjRs3qmrVqpbxd999V+vWrdPmzZszJCZ+1w4AAAAYJK1k4F5y5MghV1dXnTp1ymb81KlTCgoKyrCY6EEAAAAAHgHu7u6qUKGCVq9ebRlLTk7W6tWrbSoK/xYVBAAAAOARERkZqbZt26pixYp68sknNWbMGF29elXt27fPsDVIEJCpPDw8NGDAAJrFADvw9wZ4MPzdwX9RixYtdObMGfXv31/x8fEqW7asli9fnqpx+d+gSRkAAACABT0IAAAAACxIEAAAAABYkCAAAAAAsCBBgEOLiYmRn59fhs8FMlutWrXUs2fP+84pWLCgxowZ86/XMplMWrx4cYbPBR6mB/k3vF27dmrUqNFDiceev48Z9XcXcBQkCE7sYf7Dmh4FCxaUyWSSyWRS1qxZVapUKU2ZMsVmTosWLbR//36DIgTS1q5dO5lMJr322muprnXt2lUmk0nt2rWzjC1cuFBDhgz5V2sOHDjQ8nfG1dVV+fPnV5cuXXT+/HmbeSdPnlS9evX+1VpARknr/2vWrl0rk8mkixcvSsq8f+9jYmIsf49cXFyUJ08etWjRQnFxcTbztmzZoi5dujz0eABHRIIAQw0ePFgnT57U7t271bp1a3Xu3Fnfffed5bqXl5dy5cplYIRA2vLnz685c+bo+vXrlrGEhATNmjVLBQoUsJkbEBAgb2/vf71myZIldfLkScXFxWnatGlavny5Xn/9dZs5QUFBPNYRj5zM/Pfex8dHJ0+e1PHjx7VgwQLFxsaqWbNmNnNy5syprFmzZko8gKMhQUCa1q1bpyeffFIeHh7KkyeP+vTpo1u3bkmSli5dKj8/PyUlJUmSduzYIZPJpD59+lje36lTJ7Vu3fq+a3h7eysoKEiFCxdW7969FRAQoFWrVlmu/73k/Ntvv6l27dry9vaWj4+PKlSooF9//fWe9z5z5owqVqyoxo0bKzEx8UG/DUCaypcvr/z582vhwoWWsYULF6pAgQIqV66czdy/bzE6ffq0GjRoIC8vLxUqVEgzZ85M15pZsmRRUFCQ8uXLp/DwcDVr1szm74xku23oxo0b6tatm/LkySNPT08FBwcrKioqzfsPGDBAefLk0c6dO9MVD5BR7rXFaOjQocqVK5e8vb3VqVMn9enTR2XLlk313k8++UR58uRRYGCgunbtqps3b953LZPJpKCgIOXJk0dPPfWUOnbsqF9++UWXL1+2zLHeNmQ2mzVw4EAVKFBAHh4eyps3r3r06JHm/adMmSI/Pz+b026BRwkJAu7p+PHjev7551WpUiX99ttvmjRpkqKjozV06FBJUo0aNfTXX39p+/btklKSiRw5cmjt2rWWe6xbt061atVK13rJyclasGCBLly4IHd39zTntWrVSo899pi2bNmirVu3qk+fPnJzc0s17+jRo6pRo4aeeOIJzZ8/n9+m4qHp0KGDpk2bZvl66tSp6TrNsl27djp69Kh++OEHzZ8/XxMnTtTp06ftWvvIkSNasWLFff/OjBs3Tt98843mzp2r2NhYzZw5UwULFkw1z2w2q3v37vrf//6nH3/8UaVLl7YrFiCjzZw5U8OGDdNHH32krVu3qkCBApo0aVKqeT/88IMOHjyoH374QdOnT1dMTIxiYmLSvc7p06e1aNEiubq6ytXV9Z5zFixYoNGjR+uzzz7TgQMHtHjxYpUqVeqec0eMGKE+ffpo5cqVeuaZZ9IdB+BIOEkZ9zRx4kTlz59f48ePl8lkUmhoqE6cOKHevXurf//+8vX1VdmyZbV27VpVrFhRa9euVa9evTRo0CBduXJFly5d0h9//KGwsLD7rtO7d2/169dPiYmJunXrlgICAtSpU6c058fFxemdd95RaGioJKlo0aKp5sTGxqpOnTpq3LixxowZI5PJ9O++GcB9tG7dWn379tWff/4pSfrpp580Z84cm2T57/bv36/vvvtOv/zyiypVqiRJio6OVvHixf9xvV27dil79uxKSkpSQkKCJGnUqFFpzo+Li1PRokVVvXp1mUwmBQcHp5pz69YttW7dWtu3b9eGDRuUL1++f4wDsMfSpUuVPXt2m7E7Fei0fPrpp+rYsaMl4e7fv79WrlypK1eu2Mzz9/fX+PHj5erqqtDQUNWvX1+rV69W586d07z3pUuXlD17dpnNZl27dk2S1KNHD2XLlu2e8+Pi4hQUFKTw8HC5ubmpQIECevLJJ1PN6927t2bMmKF169apZMmS9/18gCOjgoB72rt3r6pWrWrzw3W1atV05coVHTt2TJIUFhamtWvXymw268cff1RERISKFy+uDRs2aN26dcqbN+89f4C39s4772jHjh1as2aNKleurNGjR6tIkSJpzo+MjFSnTp0UHh6uDz/8UAcPHrS5fv36ddWoUUMREREaO3YsyQEeupw5c6p+/fqKiYnRtGnTVL9+feXIkeO+79m7d6+yZMmiChUqWMZCQ0PT9QSXkJAQ7dixQ1u2bFHv3r1Vt25dde/ePc357dq1044dOxQSEqIePXpo5cqVqeb06tVLmzdv1vr160kO8FDUrl1bO3bssHn9/aEUfxcbG5vqh/B7/VBesmRJm9/858mT5x+rcd7e3tqxY4d+/fVXjRw5UuXLl9ewYcPSnN+sWTNdv35dhQsXVufOnbVo0SLLlts7Ro4cqS+++EIbNmwgOcAjjwQBD6xWrVrasGGDfvvtN7m5uSk0NFS1atXS2rVrtW7dun+sHkhSjhw5VKRIEdWoUUPz5s1Tjx49tGfPnjTnDxw4UL///rvq16+vNWvWqESJElq0aJHluoeHh8LDw7V06VIdP348Qz4n8E86dOigmJgYTZ8+XR06dHioa7m7u6tIkSJ64okn9OGHH8rV1VWDBg1Kc3758uV1+PBhDRkyRNevX1fz5s3VtGlTmzl16tTR8ePHtWLFiocaO5xXtmzZVKRIEZtXRiWjf99majKZlJycfN/3uLi4qEiRIipevLgiIyNVpUqVVM3+1vLnz6/Y2FhNnDhRXl5eeuONN1SzZk2bXocaNWooKSlJc+fO/XcfCHAAJAi4p+LFi2vTpk0ym82WsZ9++kne3t567LHHJN3tQxg9erQlGbiTIKxduzbd/Qd35M+fXy1atFDfvn3vO69YsWLq1auXVq5cqYiICJv93y4uLpoxY4YqVKig2rVr68SJE3bFADyI5557Tjdu3NDNmzdVt27df5wfGhqqW7duaevWrZax2NhYy+Me7dGvXz998skn9/3fuo+Pj1q0aKEvvvhCX331lRYsWGDzaNQXX3xRs2bNUqdOnTRnzhy7YwAehpCQEG3ZssVm7O9fZ5Q+ffroq6++0rZt29Kc4+XlpQYNGmjcuHFau3atNm3apF27dlmuP/nkk/ruu+80fPhwffLJJw8lTiCzkCA4uUuXLqUq+x49elRvvPGGjh49qu7du2vfvn36+uuvNWDAAEVGRsrFJeV/Nv7+/ipdurRmzpxpSQZq1qypbdu2af/+/emqIPzdm2++qSVLltzzyUTXr19Xt27dtHbtWv3555/66aeftGXLllT7tl1dXTVz5kyVKVNGTz/9tOLj4+3/xgB2cHV11d69e7Vnz540mxythYSE6LnnntOrr76qzZs3a+vWrerUqZO8vLzsXrtq1aoqXbq0hg8ffs/ro0aN0uzZs7Vv3z7t379f8+bNU1BQUKrtTI0bN9aMGTPUvn17zZ8/3+44gIzWvXt3RUdHa/r06Tpw4ICGDh2qnTt3PpSto/nz51fjxo3Vv3//e16PiYlRdHS0du/erUOHDunLL7+Ul5dXqp6ep556SsuWLdOgQYM4OA2PNBIEJ7d27VqVK1fO5jVo0CDly5dPy5Yt0y+//KIyZcrotddeU8eOHdWvXz+b94eFhSkpKcmSIAQEBKhEiRIKCgpSSEiI3fGUKFFCzz777D3/kXZ1ddW5c+fUpk0bFStWTM2bN1e9evXuub0iS5Ysmj17tkqWLKmnn37a7qfDAPby8fGRj49PuudPmzZNefPmVVhYmCIiItSlS5cHfgZ8r169NGXKFB09ejTVNW9vb40YMUIVK1ZUpUqVdOTIES1btsyS6Ftr2rSppk+frldeecXm0a2AEVq1aqW+ffvq7bfftmyVa9eunTw9PR/Ker169dK3336rX375JdU1Pz8/ffHFF6pWrZpKly6t77//XkuWLFFgYGCqudWrV9e3336rfv366dNPP30osQIPm8lsvYcEAADAQdWpU0dBQUGaMWOG0aEA/2k85hQAADica9euafLkyapbt65cXV01e/Zsff/996kOBgSQ8aggAAAAh3P9+nU1aNBA27dvV0JCgkJCQtSvXz9FREQYHRrwn0eCAAAAAMCCJmUAAAAAFiQIAAAAACxIEAAAAABYkCAAAAAAsCBBAAAAAGBBggAADqZdu3Zq1KiR5etatWqpZ8+emR7H2rVrZTKZdPHixUxfGwBgHBIEAEindu3ayWQyyWQyyd3dXUWKFNHgwYN169ath7ruwoULNWTIkHTN5Yd6AMC/xUnKAGCH5557TtOmTVNiYqKWLVumrl27ys3NTX379rWZd+PGDbm7u2fImgEBARlyHwAA0oMKAgDYwcPDQ0FBQQoODtbrr7+u8PBwffPNN5ZtQcOGDVPevHkVEhIiSTp69KiaN28uPz8/BQQEqGHDhjpy5IjlfklJSYqMjJSfn58CAwP17rvv6u/nV/59i1FiYqJ69+6t/Pnzy8PDQ0WKFFF0dLSOHDmi2rVrS5L8/f1lMpnUrl07SVJycrKioqJUqFAheXl5qUyZMpo/f77NOsuWLVOxYsXk5eWl2rVr28QJAHAeJAgA8C94eXnpxo0bkqTVq1crNjZWq1at0tKlS3Xz5k3VrVtX3t7e+vHHH/XTTz8pe/bseu655yzvGTlypGJiYjR16lRt2LBB58+f16JFi+67Zps2bTR79myNGzdOe/fu1Weffabs2bMrf/78WrBggSQpNjZWJ0+e1NixYyVJUVFR+t///qfJkyfr999/V69evdS6dWutW7dOUkoiExERoQYNGmjHjh3q1KmT+vTp87C+bQAAB8YWIwB4AGazWatXr9aKFSvUvXt3nTlzRtmyZdOUKVMsW4u+/PJLJScna8qUKTKZTJKkadOmyc/PT2vXrtWzzz6rMWPGqG/fvoqIiJAkTZ48WStWrEhz3f3792vu3LlatWqVwsPDJUmFCxe2XL+zHSlXrlzy8/OTlFJxGD58uL7//ntVrVrV8p4NGzbos88+U1hYmCZNmqTHH39cI0eOlCSFhIRo165d+uijjzLwuwYAeBSQIACAHZYuXars2bPr5s2bSk5O1ssvv6yBAweqa9euKlWqlE3fwW+//aY//vhD3t7eNvdISEjQwYMHdenSJZ08eVKVK1e2XMuSJYsqVqyYapvRHTt27JCrq6vCwsLSHfMff/yha9euqU6dOjbjN27cULly5SRJe/futYlDkiWZAAA4FxIEALBD7dq1NWnSJLm7uytv3rzKkuXuP6PZsmWzmXvlyhVVqFBBM2fOTHWfnDlzPtD6Xl5edr/nypUrkqRvv/1W+fLls7nm4eHxQHEAAP67SBAAwA7ZsmVTkSJF0jW3fPny+uqrr5QrVy75+Pjcc06ePHm0efNm1axZU5J069Ytbd26VeXLl7/n/FKlSik5OVnr1q2zbDGydqeCkZSUZBkrUaKEPDw8FBcXl2bloXjx4vrmm29sxn7++ed//pAAgP8cmpQB4CFp1aqVcuTIoYYNG+rHH3/U4cOHtXbtWvXo0UPHjh2TJL355pv68MMPtXjxYu3bt09vvPHGfc8wKFiwoNq2basOHTpo8eLFlnvOnTtXkhQcHCyTyaSlS5fqzJkzunLliry9vfX222+rV69emj59ug4ePKht27bp008/1fTp0yVJr732mg4cOKB33nlHsbGxmjVrlmJiYh72twgA4IBIEADgIcmaNavWr1+vAgUKKCIiQsWLF1fHjh2VkJBgqSi89dZbeuWVV9S2bVtVrVpV3t7eaty48X3vO2nSJDVt2lRvvPGGQkND1blzZ129elWSlC9fPg0aNEh9+vRR7ty51a1bN0nSkCFD9MEHHygqKkrFixfXc889p2+//VaFChWSJBUoUEALFizQ4sWLVaZMGU2ePFnDhw9/iN8dAICjMpnT6oQDAAAA4HSoIAAAAACwIEEAAAAAYEGCAAAAAMCCBAEAAACABQkCAAAAAAsSBAAAAAAWJAgAAAAALEgQAAAAAFiQIAAAAACwIEEAAAAAYEGCAAAAAMDi/88jyiWQwN99AAAAAElFTkSuQmCC\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "import pandas as pd\n", + "import seaborn as sns\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, classification_report, confusion_matrix, roc_curve, roc_auc_score\n", + "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n", + "from sklearn.neighbors import KNeighborsClassifier\n", + "from xgboost import XGBClassifier\n", + "from sklearn.preprocessing import label_binarize\n", + "\n", + "# Initialize models\n", + "models = {\n", + " 'Random Forest': RandomForestClassifier(random_state=42),\n", + " 'XGBoost': XGBClassifier(random_state=42, use_label_encoder=False, eval_metric='logloss'),\n", + " 'KNN': KNeighborsClassifier(),\n", + " 'GBM': GradientBoostingClassifier(random_state=42)\n", + "}\n", + "\n", + "# Prepare data for ROC curve\n", + "roc_data = {'Model': [], 'FPR': [], 'TPR': [], 'ROC AUC': []}\n", + "\n", + "# Metrics storage\n", + "metrics = {'Model': [], 'Accuracy': [], 'Precision': [], 'Recall': [], 'F1 Score': []}\n", + "\n", + "for name, model in models.items():\n", + " # Train model\n", + " model.fit(X_train, y_train)\n", + "\n", + " # Make predictions\n", + " y_pred = model.predict(X_test)\n", + " y_pred_prob = model.predict_proba(X_test)[:, 1] if len(model.classes_) == 2 else model.predict_proba(X_test)\n", + "\n", + " # Calculate metrics\n", + " accuracy = accuracy_score(y_test, y_pred)\n", + " precision = precision_score(y_test, y_pred, average='weighted')\n", + " recall = recall_score(y_test, y_pred, average='weighted')\n", + " f1 = f1_score(y_test, y_pred, average='weighted')\n", + "\n", + " # ROC curve data\n", + " if len(model.classes_) == 2: # Binary classification\n", + " fpr, tpr, _ = roc_curve(y_test, y_pred_prob)\n", + " roc_auc = roc_auc_score(y_test, y_pred_prob)\n", + " else: # Multi-class classification\n", + " y_test_bin = label_binarize(y_test, classes=model.classes_)\n", + " fpr, tpr, _ = roc_curve(y_test_bin.ravel(), model.predict_proba(X_test).ravel())\n", + " roc_auc = roc_auc_score(y_test_bin, model.predict_proba(X_test), average='weighted', multi_class='ovr')\n", + "\n", + " # Store metrics\n", + " metrics['Model'].append(name)\n", + " metrics['Accuracy'].append(accuracy)\n", + " metrics['Precision'].append(precision)\n", + " metrics['Recall'].append(recall)\n", + " metrics['F1 Score'].append(f1)\n", + "\n", + " roc_data['Model'].append(name)\n", + " roc_data['FPR'].append(fpr)\n", + " roc_data['TPR'].append(tpr)\n", + " roc_data['ROC AUC'].append(roc_auc)\n", + "\n", + "# Convert metrics and ROC data to DataFrames\n", + "metrics_df = pd.DataFrame(metrics)\n", + "roc_df = pd.DataFrame(roc_data)\n", + "\n", + "# Plot Count Plots for Accuracy, Precision, Recall, and F1 Score\n", + "metrics_melted = metrics_df.melt(id_vars='Model', var_name='Metric', value_name='Score')\n", + "\n", + "plt.figure(figsize=(14, 10))\n", + "sns.barplot(data=metrics_melted, x='Model', y='Score', hue='Metric')\n", + "plt.title('Model Performance Metrics')\n", + "plt.xlabel('Model')\n", + "plt.ylabel('Score')\n", + "plt.legend(title='Metric')\n", + "plt.show()\n", + "\n", + "# Plot ROC Curves\n", + "plt.figure(figsize=(14, 10))\n", + "for idx, row in roc_df.iterrows():\n", + " plt.plot(row['FPR'], row['TPR'], label=f\"{row['Model']} (AUC = {row['ROC AUC']:.2f})\")\n", + "\n", + "plt.plot([0, 1], [0, 1], 'k--')\n", + "plt.xlabel('False Positive Rate')\n", + "plt.ylabel('True Positive Rate')\n", + "plt.title('ROC Curve')\n", + "plt.legend()\n", + "plt.show()\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "stBfnk1qlGz0", + "outputId": "6419c5ac-fe6f-4f0a-d554-0cf30ac45552" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "import joblib\n", + "# Assuming model_xgb is your trained model\n", + "joblib.dump(model_xgb, 'xgb_model.pkl')" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "j1WYuO36mQTu", + "outputId": "f6b83223-832e-40b5-e536-e734254f4035" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "['xgb_model.pkl']" + ] + }, + "metadata": {}, + "execution_count": 51 + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Load the trained XGBoost model\n", + "model_xgb = joblib.load('xgb_model.pkl')\n", + "\n", + "# Function to make predictions\n", + "def predict_risk_level(age, systolic_bp, diastolic_bp, bs, body_temp, heart_rate):\n", + " # Create a DataFrame for the input data\n", + " input_data = pd.DataFrame({\n", + " 'Age': [age],\n", + " 'SystolicBP': [systolic_bp],\n", + " 'DiastolicBP': [diastolic_bp],\n", + " 'BS': [bs],\n", + " 'BodyTemp': [body_temp],\n", + " 'HeartRate': [heart_rate]\n", + " })\n", + "\n", + " # Predict using the loaded model\n", + " prediction = model_xgb.predict(input_data)\n", + "\n", + " # Map prediction to risk level\n", + " risk_levels = ['low Maternal risk', 'Medium Maternal risk', 'High Maternal risk']\n", + " return risk_levels[prediction[0]]\n", + "\n", + "# Interactive input cells\n", + "age = int(input('Enter Age: '))\n", + "systolic_bp = int(input('Enter Systolic BP: '))\n", + "diastolic_bp = int(input('Enter Diastolic BP: '))\n", + "bs = float(input('Enter Blood Sugar: '))\n", + "body_temp = float(input('Enter Body Temperature (Fahrenheit): '))\n", + "heart_rate = int(input('Enter Heart Rate: '))\n", + "\n", + "# Predict risk level\n", + "risk_level = predict_risk_level(age, systolic_bp, diastolic_bp, bs, body_temp, heart_rate)\n", + "print(f'Predicted Risk Level: {risk_level}')\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "5VuyYPo2oXPl", + "outputId": "fe3970ba-a2bc-4c00-dcae-a7c2a8fab62c" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Enter Age: 67\n", + "Enter Systolic BP: 160\n", + "Enter Diastolic BP: 90\n", + "Enter Blood Sugar: 15\n", + "Enter Body Temperature (Fahrenheit): 99\n", + "Enter Heart Rate: 89\n", + "Predicted Risk Level: low Maternal risk\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "import pandas as pd\n", + "import joblib\n", + "\n", + "# Load the trained XGBoost model\n", + "model_xgb = joblib.load('xgb_model.pkl')\n", + "\n", + "def predict_risk_level(age, systolic_bp, diastolic_bp, bs, body_temp, heart_rate):\n", + " # Create a DataFrame for the input data\n", + " input_data = pd.DataFrame({\n", + " 'Age': [age],\n", + " 'SystolicBP': [systolic_bp],\n", + " 'DiastolicBP': [diastolic_bp],\n", + " 'BS': [bs],\n", + " 'BodyTemp': [body_temp],\n", + " 'HeartRate': [heart_rate]\n", + " })\n", + "\n", + " # Predict using the loaded model\n", + " prediction = model_xgb.predict(input_data)\n", + "\n", + " # Map prediction to risk level\n", + " if prediction[0] == 0:\n", + " risk_level = 'Low Maternal risk'\n", + " elif prediction[0] == 1:\n", + " risk_level = 'Medium Maternal risk'\n", + " elif prediction[0] == 2:\n", + " risk_level = 'High Maternal risk'\n", + "\n", + " return risk_level\n", + "\n", + "# Test with example values\n", + "print(predict_risk_level(30, 120, 80, 7.0, 98.6, 80))\n", + "print(predict_risk_level(50, 140, 90, 22.0, 100.0, 70))\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "FX9-rRN3xfvB", + "outputId": "161c0c93-4e32-462e-bce8-2389e7bb811b" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Low Maternal risk\n", + "Low Maternal risk\n" + ] + } + ] + } + ] +} \ No newline at end of file diff --git a/Maternal health Risk Prediction/readme.md b/Maternal health Risk Prediction/readme.md new file mode 100644 index 00000000..6cd2d987 --- /dev/null +++ b/Maternal health Risk Prediction/readme.md @@ -0,0 +1,84 @@ +# Maternal Health Risk prediction using machine learning + +The Maternal Risk Prediction project is a Streamlit web application designed to predict the risk level (low, medium, or high) for maternal health based on specific health metrics. The application utilizes a trained XGBoost model to make predictions. Users input their age, systolic blood pressure, diastolic blood pressure, blood sugar levels, body temperature, and heart rate. The model processes these inputs and provides a risk assessment. The interface is user-friendly, featuring a sleek design with a blurred background image for aesthetic appeal and clarity. This application aims to offer a quick and accessible tool for expecting mothers to assess their health risks, facilitating better-informed healthcare decisions. + +## Data Set + +The below csv dataset from kaggle is used as reference which contains nearly 1000+ rows (maternal Health risk data Set.csv) on which porcessing is performed to obtained a processed data. + +The dataset link is are as follows :- +https://www.kaggle.com/datasets/csafrit2/maternal-health-risk-data + +on this dataset, below processing are performed : +1) featue scaling and column reinitialization +2) errors and outliers removal +3) remove na,missing values , regularization etc +(all this works ar depicted in .ipynb file) + +The model is trained on processed_data_car.csv file and all works associated with it are depicted in +maternal_health_risk_prediction.ipynb file. + +## Methodology + +The project follows the below structured methodology ranging from data preprocessing pipeline to model training, evaluation and deployment :- + +1. **Data Preprocessing and feature enginnering**: +2. **Exploratory Data Analysis (EDA)**: + after Data preprocessing the next step is Exploratory data analysis using different plotting libraries like matplotlib,pandas,seaborn and plotly.following plots were plotted in this step:- + 1) Pie chart + 2) violen plot + 3) box plot of numerical features + 4) count plot + 5) heatmap or confusion matrix for four different models of machine learning + 6) model comparison graphs + 7) line plots + (refer images folder for this images and graph observation) + + +4. **Model Training and evaluation**: + The four machine learning model random forest ,xgboost ,KNN, gradient boosting machine are selected for model training over the inputed processed data: + random forest accuracy : 85 % + GBM accuracy : 77 % + XGboost accuracy : 86 % + KNN accuracy : 76 % + + The XGBoost machine model is then loaded into streamlit application after installing and using joblib library. + +5. **Inference**: + Deployed the model with the help streamlit web application to predict the maternal health risk associated with a women. + + +## Libraries Used + +1. **Joblib**: For downloading the random forest model +2. **Scikit learn**: For machine learning processing and operations +3. **Matplotlib**: For plotting and visualizing the detection results. +4. **Pandas**: For image manipulation. +5. **NumPy**: For efficient numerical operations. +6. **Seaborn** : for advanced data visualizations +7. **plotly** : for 3D data visualizations . +8. **Streamlit** : for creating gui of the web application. + + +## How to Use + +1. **Clone the Repository**: + ```sh + git clone url_to_this_repository + ``` + +2. **Install Dependencies**: + ```sh + pip install -r requirements.txt + ``` + +3. **Run the Model**: + ```python + streamlit run app.py + ``` + +4. **View Results**: The script will allow you to predict the maternal risk associated with the lady in pregnancy and therby taking care to avoid it. + + +## Demo : + diff --git a/Maternal health Risk Prediction/requirement.txt b/Maternal health Risk Prediction/requirement.txt new file mode 100644 index 00000000..9e602176 --- /dev/null +++ b/Maternal health Risk Prediction/requirement.txt @@ -0,0 +1,6 @@ +pandas==2.0.3 +numpy==1.25.2 +scikit-learn==1.2.2 +joblib==1.4.2 +streamlit +xgboost \ No newline at end of file diff --git a/Maternal health Risk Prediction/xgb_model.pkl b/Maternal health Risk Prediction/xgb_model.pkl new file mode 100644 index 00000000..96d724a3 Binary files /dev/null and b/Maternal health Risk Prediction/xgb_model.pkl differ