Structured data analysis 1: lymphoma - Duplicate

# Run this cell to start. import numpy as np # Load the OKpy test library and tests. from client.api.notebook import Notebook ok = Notebook('lymphoma.ok')

#- Simulate 10000 trials of 17 patients n_patients=18 p_cured= 0.9 counts = np.array([]) for i in np.arange(0,10000): trial= np.random.uniform(0,2,n_patients) cured= np.count_nonzero(trial <p_cured) counts= np.append(counts, cured) # Show the first five counts counts[:5]

# Test you are on the right track. _ = ok.grade('q_01_counts')

p_100 = (np.count_nonzero(counts == 17))/10000 # Show the value p_100

# Test you are on the right track. _ = ok.grade('q_02_p_100')

# This code is rather ugly, and needs editing to work correctly. n_patients =17 p_cured=0.9 labs=4 lab_counts= np.array([]) for i in np.arange(0,labs): trial= np.random.uniform(0,1,n_patients) cured= np.count_nonzero(trial< p_cured) lab_counts= np.append(lab_counts,cured) lab_counts = np.zeros(4) print(lab_counts[0], "patients cured in the first lab") # Simulate a count of patients cured in first lab. print(lab_counts[1], "patients cured in the second lab") # Simulate a count of patients cured in second lab. print(lab_counts[2], "patients cured in the third lab") # Simulate a count of patients cured in third lab. print(lab_counts[3], "patients cured in the fourth lab") # Simulate a count of patients cured for fourth lab. n_publications = np.count_nonzero(lab_counts == 17/labs) print(n_publications, "proability to publish")

#- Simulate 10000 trials of four labs, each studying 17 patients. publications = np.array ([]) trial= np.array([]) cured= np.array ([]) n_publications_2= np.array([]) for i in np.arange(0,10000): lab_counts_2= np.array([]) for j in np.arange(0, labs): trial= np.random.uniform(0,1,n_patients) cured= np.count_nonzero(trial< p_cured) lab_counts_2= np.append(lab_counts_2, cured) n_publications_2= np.count_nonzero(lab_counts_2==n_patients) publications= np.append(publications, n_publications_2) # Show the first five publication counts publications[:5]

# Test you are on the right track. _ = ok.grade('q_04_publications')

p_at_least_one = (np.count_nonzero(publications >= 1)/10000) # Show the value p_at_least_one

# Test you are on the right track. _ = ok.grade('q_05_p_at_least_one')

# For your convenience, you can run this cell to run all the tests at once! import os _ = [ok.grade(q[:-3]) for q in os.listdir("tests") if q.startswith('q')]