import numpy as np import pandas as pd from lifelines import KaplanMeierFitter from matplotlib import pyplot as plt data= pd.read_csv('C:\work\data1.csv') data.describe() data.isnull().sum()
T = data['Delay'] #survival time in days E= data['Event'] # status plt.figure(figsize=(15,8)) plt.hist(T,bins=50) plt.show()
kmf = KaplanMeierFitter() kmf.fit(durations = T, event_observed = E) kmf.plot_survival_function() plt.ylabel("probability of survival") what is the probability that person lives upto 2000 days is 0.5!
plt.title("survival curve")
plt.show()
kmf.survival_function_.plot() same plot without the 95% confidence interval
plt.title('Survival function')
plt.figure(figsize=(10,5)) ax = plt.subplot(111) m = (data["gender"] == 1) kmf.fit(durations = T[m], event_observed = E[m], label = "Male") kmf.plot_survival_function(ax = ax) kmf.fit(T[~m], event_observed = E[~m], label = "Female") kmf.plot_survival_function(ax = ax, at_risk_counts = True)
plt.title("Survival of different gender group")
plt.figure(figsize=(10,5))
ax = plt.subplot(111)
A = (data["race"] == "ASIAN") kmf.fit(durations = T[A], event_observed = E[A], label = "ASIANS") kmf.plot_survival_function(ax = ax)
W = (data["race"] == "WHITE") kmf.fit(T[W], event_observed = E[W], label = "WHITE") kmf.plot_survival_function(ax = ax)
B = (data["race"] == "BLACK OR AFRICAN AMERICAN") kmf.fit(T[B], event_observed = E[B], label = "BLACK OR AFRICAN AMERICAN") kmf.plot_survival_function(ax = ax)
AI = (data["race"] == "AMERICAN INDIAN OR ALASKA Native") kmf.fit(T[AI], event_observed = E[AI], label = "AMERICAN INDIAN OR ALASKA Native") kmf.plot_survival_function(ax = ax)
H = (data["race"] == "Native HAWAIIAN OR OTHER PACIFIC ISLANDER") kmf.fit(T[H], event_observed = E[H], label = "Native HAWAIIAN OR OTHER PACIFIC ISLANDER") kmf.plot_survival_function(ax = ax)
plt.title("Survival of different Race group")
sns.set(style= 'darkgrid') plt.figure(figsize=(10,8)) his= sns.distplot(data["age_at_initial_pathologic_diagnosis"],color = 'b') plt.title("age_at_initial_pathologic_diagnosis", fontsize= "14") plt.ylabel('FREQUENCIES', color= "b", fontsize= "14") plt.xlabel('AGE', color = "b", fontsize= "14")
