Master

PythonAssignment2.py

@@ -0,0 +1,97 @@
+import datetime
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+from matplotlib import style
+style.use('ggplot')
+
+hospitaldata = pd.read_csv(r'D:\DIH\Assignments\FarooqUrRehman_KCI_Python_Assignment2\hospitaldata.csv')
+
+#Answer 1
+hospitaldata = hospitaldata.rename(columns=lambda x: x.replace('.', ''))
+print(hospitaldata.columns)
+
+#Answer 2
+hospitaldata['WeekDay'] = pd.to_datetime(hospitaldata['Date']).dt.weekday_name
+WeekDay_Counts=hospitaldata[['WeekDay','id']].groupby('WeekDay').agg('count')
+WeekDay_Max = WeekDay_Counts['id'].max()
+print(WeekDay_Counts[(WeekDay_Counts['id'] == WeekDay_Max)])
+
+#Answer 3
+
+Average_Age = hospitaldata.dropna(subset =['Age'])
+Average_Age = Average_Age[(Average_Age['Age']!='-')]
+Average_Age['Age_Years']=Average_Age['Age'].str.extract('(\d+)').astype(int)
+Average_Age['Age_Years']=np.where(Average_Age['Age'].str.contains('M') == True, Average_Age['Age_Years']/12.0, Average_Age['Age_Years'])
+print(Average_Age['Age_Years'].mean())
+
+#Answer 4
+Children_Count = Average_Age[(Average_Age['Age_Years'] >=1) & (Average_Age['Age_Years'] <=12)]
+print(Children_Count['Age_Years'].count())
+
+#Answer 5
+Gender_Procedure = hospitaldata.dropna(subset =['Sex'])
+Gender_Procedure = Gender_Procedure.dropna(subset =['Procedure'])
+Gender_Procedure = Gender_Procedure[(Gender_Procedure['Sex']!='-')]
+Gender_Procedure['Sex']=Gender_Procedure['Sex'].str.replace("m","M")
+Gender_Procedure['Sex']=Gender_Procedure['Sex'].str.replace("f","F")
+print(Gender_Procedure.groupby('Sex')['Procedure'].value_counts().nlargest(2))
+
+#Answer 6
+Doctor_Earnings = hospitaldata[(hospitaldata['ConsultingDoctor'].str.contains("Dr")==True)]
+print(Doctor_Earnings.groupby("ConsultingDoctor")['AmountReceived'].max().nlargest(1))
+
+#Answer 7
+Procedure_Earnings=hospitaldata.dropna(subset =['Procedure'])
+Procedure_Earnings=hospitaldata.dropna(subset =['AmountReceived'])
+print(Procedure_Earnings.groupby("Procedure")['AmountReceived'].sum().nlargest(1))
+
+#Answer 8
+VF = hospitaldata.dropna(subset =['Time'])
+VF = VF[(VF['Time']!='-')]
+#Visit_Frequency['AM_PM']=np.where(Average_Age['Time'].str.contains('PM') == True, "PM","AM")
+VF['VisitHour']=pd.to_datetime(VF['Time'])
+VF['VisitHour']=VF['VisitHour'].dt.hour
+VF1=VF[['VisitHour','id']]
+VF1.groupby('VisitHour')['id'].count().nlargest(1)
+
+#Answer 9
+VF['TimeBracket']= np.where((VF['VisitHour'] >=6) & (VF['VisitHour'] <12), "Morning",np.where((VF['VisitHour'] >=12) & (VF['VisitHour'] <16), "Afternoon",np.where((VF['VisitHour'] >=16) & (VF['VisitHour'] <19), "Evening","Night")))
+VF[['TimeBracket','VisitHour']]
+
+#Answer 10 & 11
+VF2=VF[['id','Date']].groupby('id').count()
+VF2[VF2['Date'] > 1]
+
+# Answer 12
+Gender_Procedure2=Gender_Procedure[['id','Procedure','Date']].groupby(['id','Procedure']).count()
+Gender_Procedure2[Gender_Procedure2['Date']>1]
+
+# Answer 13
+Average_Age['Sex']=Average_Age['Sex'].str.replace("m","M")
+Average_Age['Sex']=Average_Age['Sex'].str.replace("f","F")
+Average_Age.groupby("Sex")['Age_Years'].mean()
+
+# Answer 14
+Gender_Procedure = Gender_Procedure.dropna(subset =['AmountBalance'])
+Gender_Procedure = Gender_Procedure[(Gender_Procedure['AmountBalance']!=' -   ')]
+Gender_Procedure['AmountBalance1'] = Gender_Procedure['AmountBalance'].str.replace(',','')
+Gender_Procedure['AmountBalance1'] = Gender_Procedure['AmountBalance1'].str.extract('(\d+)').astype(float)
+sum(Gender_Procedure['AmountBalance1'])
+
+# Answer 15
+Consultation_Earnings=Procedure_Earnings[(Procedure_Earnings['Procedure'] == 'Consultation')]
+print(sum(Consultation_Earnings['AmountReceived']))
+
+# Answer 16
+Age_Charges_Cor = Average_Age.dropna(subset =['TotalCharges'])
+Age_Charges_Cor = Age_Charges_Cor[(Age_Charges_Cor['TotalCharges']!='Cancelled')]
+np.correlate(Age_Charges_Cor['Age_Years'],Age_Charges_Cor['TotalCharges'].str.extract('(\d+)').astype(float))
+
+# Answer 17
+Average_Age['Age_Group']= np.where((Average_Age['Age_Years'] <=15), "1-15 Yrs",np.where((Average_Age['Age_Years'] >15) & (Average_Age['Age_Years'] <=30), "16-30 Yrs",np.where((Average_Age['Age_Years'] >30) & (Average_Age['Age_Years'] <=45), "31-45 Yrs","Above 45 Yrs")))
+print(Average_Age.groupby('Age_Group')['id'].count())
+
+# Answer 18
+XRay_Scalling_Earnings=Procedure_Earnings[(Procedure_Earnings['Procedure'] == 'X Ray') | (Procedure_Earnings['Procedure'] == 'Scalling')]
+print(sum(XRay_Scalling_Earnings['AmountReceived']))

RAssignment_2.R

@@ -0,0 +1,158 @@
+library(tidyr)
+library(lubridate)
+library(dplyr)
+library(plyr)
+
+#hospitaldata <- read.csv("D://DIH//Assignments//FarooqUrRehman_KCI_R_Assignment2//hospitaldata.csv")
+
+## Answer 1
+names(hospitaldata) <- gsub("\\.", "", names(hospitaldata))
+
+## Answer 2
+day_count <- count(wday(mdy(hospitaldata$Date),label=TRUE))
+day_max <- max(day_count$freq)
+select(filter(day_count, day_count$freq == day_max), x)
+
+## Answer 3
+hospitaldata <- hospitaldata%>%
+  mutate(Age=ifelse(Age=='-', NA, Age))%>%
+  mutate(Age=ifelse(grepl("M",Age),parse_number(Age)/12,Age))
+
+print(mean(as.numeric(hospitaldata$Age),na.rm = TRUE))
+
+##Answer 4
+hospitaldata <- hospitaldata %>%
+  mutate(Age = as.numeric(Age)) 
+
+children_data <- hospitaldata %>%
+  filter(Age >= 1, Age <=12) %>%
+  select()%>%
+  count() %>%
+  sum() %>%
+  print
+
+# Answer 5
+gen_proc<-hospitaldata %>%
+  mutate(Sex = ifelse(Sex == 'M' | Sex =='m', 'M',ifelse(Sex == 'f' | Sex =='F','F',NA))) %>%
+  filter(!is.na(Sex), !is.na(Procedure), Sex == 'F') %>%
+  select(Procedure) %>%
+  count()
+gen_proc_1 <- max(gen_proc$freq)
+select(filter(gen_proc,gen_proc$freq == gen_proc_1),Procedure,freq)
+
+gen_proc<-hospitaldata %>%
+  mutate(Sex = ifelse(Sex == 'M' | Sex =='m', 'M',ifelse(Sex == 'f' | Sex =='F','F',NA))) %>%
+  filter(!is.na(Sex), !is.na(Procedure), Sex == 'M') %>%
+  select(Procedure) %>%
+  count()
+gen_proc_1 <- max(gen_proc$freq)
+select(filter(gen_proc,gen_proc$freq == gen_proc_1),Procedure,freq)
+
+
+# Answer 6
+doctors_earnings <- hospitaldata %>%
+  filter(!is.na(AmountReceived),grepl("^Dr",ConsultingDoctor)) %>%
+  group_by(ConsultingDoctor) %>%
+  summarise(sum(AmountReceived))
+
+max_earning <- max(doctors_earnings$`sum(AmountReceived)`)
+print (filter(doctors_earnings, doctors_earnings$`sum(AmountReceived)` == max_earning))
+
+# Answer 7
+proc_earnings <- hospitaldata %>%
+  filter(!is.na(AmountReceived),!is.na(Procedure)) %>%
+  group_by(Procedure) %>%
+  summarise(sum(AmountReceived))
+
+max_proc_earning <- max(proc_earnings$`sum(AmountReceived)`)
+print (filter(proc_earnings, proc_earnings$`sum(AmountReceived)` == max_proc_earning))
+
+# Answer 8
+
+cons_time <- hospitaldata %>%
+  filter(!is.na(Time),!grepl("-",Time)) %>%
+  mutate(am_pm=ifelse(grepl("AM",Time),"AM",ifelse(grepl("PM",Time),"PM","AM"))) %>%
+  group_by(cons_time=paste(as.character(hour(hm(Time))),am_pm, sep=":00 ")) %>%
+  tally()
+
+cons_max_count <- max(cons_time$n)
+print (filter(cons_time, cons_time$n == cons_max_count))
+
+# Answer 9
+Time_Bracket <- hospitaldata %>%
+  filter(!is.na(Time),!grepl("-",Time)) %>%
+  mutate(CHour = as.numeric(hour(hm(Time)))) %>%
+  mutate(CHour = ifelse(grepl("AM",Time) & CHour == 12,0,ifelse(grepl("PM",Time) & CHour<12,12+CHour,CHour))) %>%
+  mutate(TimeBracket = ifelse(CHour >=6 & CHour<12,"Morning",ifelse(CHour >=12 & CHour<16,"Afternoon",ifelse(CHour >=16 & CHour<19,"Evening","Night"))))%>%
+  print(Time,TimeBracket)
+
+# Answer 10 and 11
+Repeated_Visitors <- hospitaldata %>%
+  filter(grepl("^Dr",ConsultingDoctor)) %>%  
+  group_by(id) %>%
+  tally()%>%
+  filter(n>1) %>%
+  print 
+
+# Answer 12
+Rep_with_same_problem <- hospitaldata %>%
+  filter(grepl("^Dr",ConsultingDoctor)) %>%
+  group_by(id, Procedure) %>%
+  tally()%>%
+  filter(n>1) %>%
+  print
+
+# Answer 13
+median_age <- hospitaldata %>%
+  filter(!is.na(Sex), !grepl("-",Sex), !is.na(Age)) %>%
+  group_by(Sex) %>%
+  summarise_each(funs(median),Age) %>%
+  print
+
+# Answer 14
+balance_amount <-hospitaldata %>%
+  filter(!is.na(AmountBalance), !grepl("-",AmountBalance)) %>%
+  mutate(AmountBalance=parse_number(AmountBalance)) %>%
+  summarise(sum(AmountBalance)) %>%
+  print
+
+# Answer 15
+Consultation_amount <-hospitaldata %>%
+  filter(!is.na(AmountReceived), Procedure=='Consultation') %>%
+  summarise(sum(AmountReceived)) %>%
+  print
+
+# Answer 15
+Consultation_amount <-hospitaldata %>%
+  filter(!is.na(AmountReceived), Procedure=='Consultation') %>%
+  summarise(sum(AmountReceived)) %>%
+  print
+
+# Answer 16
+Age_n_Charges_Relation <- hospitaldata %>%
+  filter(!is.na(Age), !is.na(TotalCharges), !grepl("Cancelled",TotalCharges)) %>%
+  mutate(TotalCharges = parse_number(TotalCharges)) %>%
+  select(TotalCharges, Age) %>%
+  cor() %>%
+  print
+
+# Answer 17
+AgeGroup_Visits <-hospitaldata %>%
+  filter(!is.na(Age)) %>%
+  mutate(Age_Group=ifelse(Age<=10,'0 - 10 yrs',ifelse(Age>10 & Age <= 30,'11 - 30 yrs',ifelse(Age>31 & Age <= 50,'31 - 50 yrs','51 yrs and onwards'))))%>%
+  group_by(Age_Group)%>%
+  tally() %>%
+  print
+
+# Answer 18
+XRay_Scalling_Cost <-hospitaldata %>%
+  filter(!is.na(AmountReceived), grepl("X Ray",Procedure) | grepl("Scalling",Procedure)) %>%
+  summarise(sum(AmountReceived))%>%
+  print
+
+# 2nd answer
+XRay_Scalling_Cost <-hospitaldata %>%
+  filter(!is.na(AmountReceived), Procedure=="X Ray" | Procedure == "Scalling") %>%
+  summarise(sum(AmountReceived)) %>%
+  print
+