CredalCAT/cat.py at main · alessandroantonucci/CredalCAT · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
import os
import matplotlib.pyplot as plt
import pandas as pd
import csv
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import style
import matplotlib.patches as mpatches
from tikzplotlib import save as tikz_save

#directory = os.listdir(".")
#for filename in directory:
#	print(filename)
n_students = 1024
n_questions = 18

profiles_csv = 'Minimalistic1x2x9.profiles.csv'
profiles = pd.read_csv(profiles_csv,names=['pr'])

head_c = ['s','skill','q','t','sc','a','l0','l1','u0','u1']
head_b = ['s','skill','q','t','sc','a','p0','p1']

c_mode_csv = 'Minimalistic1x2x9.progress.credal-adaptive-mode.eps01.csv'
c_ent_csv = 'Minimalistic1x2x9.progress.credal-adaptive-entropy.eps01.csv'
c_mode2_csv = 'Minimalistic1x2x9.progress.credal-adaptive-mode.csv'
c_ent2_csv = 'Minimalistic1x2x9.progress.credal-adaptive-entropy.csv'
b_ran_csv = 'Minimalistic1x2x9.progress.bayesian-non-adaptive.csv'
b_mode_csv = 'Minimalistic1x2x9.progress.bayesian-adaptive-mode.csv'
b_ent_csv = 'Minimalistic1x2x9.progress.bayesian-adaptive-entropy.csv'
b_pr_csv = 'Minimalistic1x2x9.progress.bayesian-adaptive-pright.csv'
c_pr_csv = 'Minimalistic1x2x9.progress.credal-adaptive-pright.csv'

c_mode = pd.read_csv(c_mode_csv,names=head_c)
c_ent = pd.read_csv(c_ent_csv,names=head_c)
c_mode2 = pd.read_csv(c_mode2_csv,names=head_c)
c_ent2 = pd.read_csv(c_ent2_csv,names=head_c)
c_pr = pd.read_csv(c_pr_csv,names=head_c)
b_ran = pd.read_csv(b_ran_csv,names=head_b)
b_mode = pd.read_csv(b_mode_csv,names=head_b)
b_ent = pd.read_csv(b_ent_csv,names=head_b)
b_pr = pd.read_csv(b_pr_csv,names=head_b)

nn1 = np.zeros(n_questions+1)
nn2 = np.zeros(n_questions+1)
nn3 = np.zeros(n_questions+1)
nn4 = np.zeros(n_questions+1)
nn5 = np.zeros(n_questions+1)
nn6 = np.zeros(n_questions+1)
nn7 = np.zeros(n_questions+1)
nn8 = np.zeros(n_questions+1)
nn9 = np.zeros(n_questions+1)

br1 = np.zeros(n_questions+1)
br2 = np.zeros(n_questions+1)
br3 = np.zeros(n_questions+1)
br4 = np.zeros(n_questions+1)
br5 = np.zeros(n_questions+1)
br6 = np.zeros(n_questions+1)
br7 = np.zeros(n_questions+1)
br8 = np.zeros(n_questions+1)
br9 = np.zeros(n_questions+1)

x = range(n_questions+1)

for s in range(n_students):
    p = profiles['pr'][s]
    d1 = c_ent[c_ent['s']==s]
    d2 = c_ent2[c_ent2['s']==s]
    d3 = c_mode[c_mode['s']==s]
    d4 = c_mode2[c_mode2['s']==s]
    d5 = c_pr[c_pr['s']==s]
    d6 = b_ran[b_ran['s']==s]
    d7 = b_ent[b_ent['s']==s]
    d8 = b_mode[b_mode['s']==s]
    d9 = b_pr[b_pr['s']==s]
    #x1 = range(len(d1['l0']))
    #x2 = range(len(d2['l0']))
    #x3 = range(len(d3['l0']))
    #x4 = range(len(d4['l0']))
    #x5 = range(len(d5['l0']))
    #x6 = range(len(d6['p0']))
    #x7 = range(len(d7['p0']))
    #x8 = range(len(d8['p0']))
    #x9 = range(len(d9['p0']))
    l1 = np.array(d1['l0'] if p else d1['l1'])
    u1 = np.array(d1['u0'] if p else d1['u1'])
    l2 = np.array(d2['l0'] if p else d2['l1'])
    u2 = np.array(d2['u0'] if p else d2['u1'])
    l3 = np.array(d3['l0'] if p else d3['l1'])
    u3 = np.array(d3['u0'] if p else d3['u1'])
    l4 = np.array(d4['l0'] if p else d4['l1'])
    u4 = np.array(d4['u0'] if p else d4['u1'])
    l5 = np.array(d5['l0'] if p else d5['l1'])
    u5 = np.array(d5['u0'] if p else d5['u1'])
    p6 = np.array(d6['p0'] if p else d6['p1'])
    p7 = np.array(d7['p0'] if p else d7['p1'])
    p8 = np.array(d8['p0'] if p else d8['p1'])
    p9 = np.array(d9['p0'] if p else d9['p1'])
    u1 = np.append(u1,[u1[-1] for _ in range(n_questions+1-len(u1))])
    u2 = np.append(u2,[u2[-1] for _ in range(n_questions+1-len(u2))])
    u3 = np.append(u3,[u3[-1] for _ in range(n_questions+1-len(u3))])
    u4 = np.append(u4,[u4[-1] for _ in range(n_questions+1-len(u4))])
    l1 = np.append(l1,[l1[-1] for _ in range(n_questions+1-len(l1))])
    l2 = np.append(l2,[l2[-1] for _ in range(n_questions+1-len(l2))])
    l3 = np.append(l3,[l3[-1] for _ in range(n_questions+1-len(l3))])
    l4 = np.append(l4,[l4[-1] for _ in range(n_questions+1-len(l4))])
    l5 = np.append(l5,[l5[-1] for _ in range(n_questions+1-len(l5))])
    u5 = np.append(u5,[u5[-1] for _ in range(n_questions+1-len(u5))])
    p6 = np.append(p6,[p6[-1] for _ in range(n_questions+1-len(p6))])
    p7 = np.append(p7,[p7[-1] for _ in range(n_questions+1-len(p7))])
    p8 = np.append(p8,[p8[-1] for _ in range(n_questions+1-len(p8))])
    p9 = np.append(p9,[p9[-1] for _ in range(n_questions+1-len(p9))])

    #nn1 += (l1<0.5)
    #nn3 += (l3<0.5)
    nn1 += ((l1+u1)<1.)
    nn3 += ((l3+u3)<1.)
    nn6 += (p6<0.5)
    nn7 += (p7<0.5)
    nn8 += (p8<0.5)
    nn9 += (p9<0.5)
    br1 += l1#+u1)
    br3 += l3#+u3)
    br6 += p6
    br7 += p7
    br8 += p8
    br9 += p9

    if False:
        print(np.array(d1['t']))
        print(np.array(d2['t']))
        print(np.array(d3['t']))
        print(np.array(d4['t']))
        print(np.array(d5['t']))
        print(np.array(d6['t']))
        print(np.array(d7['t']))
        print(np.array(d8['t']))
        print(np.array(d9['t']))

    if False:
        plt.fill_between(x, l1, u1, alpha=0.1, color='red',label='H')
        plt.fill_between(x, l3, u3, alpha=0.1, color='blue',label='mode')
        plt.fill_between(x,l5,u5,alpha=0.2,color='yellow')
        plt.xlabel("Questions")
        plt.ylabel("P(S)")
        plt.xticks(np.arange(n_questions+1))
        plt.plot(x,p6,'--k',label='rand')
        plt.plot(x,p7,'-r',label='H')
        plt.plot(x,p8,'-b',label='mode')
        plt.legend()
        plt.show()

nn1 /= n_students
nn3 /= n_students
nn6 /= n_students
nn7 /= n_students
nn8 /= n_students
nn9 /= n_students
br1 /= n_students
br3 /= n_students
br6 /= n_students
br7 /= n_students
br8 /= n_students
br9 /= n_students

#plt.yscale("log")


#plt.plot(range(len(totals_precise)), totals_precise, color='black', label='ciao')
#plt.fill_between(range(len(totals_precise)), totals_lower, totals_upper, color='black', alpha=.2)
#plt.xlabel("$t$")
#plt.ylabel("$cost$")
# plt.yticks(np.arange(0, 1.01, step=0.25))
fig = plt.figure()
plt.plot(np.arange(n_questions+1), nn1,'r', label='c-h')
plt.plot(np.arange(n_questions+1), nn3,'b', label='c-m')
plt.plot(np.arange(n_questions+1), nn6,'k--', label='rand')
plt.plot(np.arange(n_questions+1), nn7,'r--', label='h')
plt.plot(np.arange(n_questions+1), nn8,'b--', label='m')
plt.plot(np.arange(n_questions+1), nn9,'y--', label='r')
plt.legend()
#plt.show()
plt.savefig("acc.png", dpi=250, bbox_inches="tight",pad_inches=0.02)
tikz_save('acc.tex')

plt.plot(np.arange(n_questions+1), br1,'r', label='c-h')
plt.plot(np.arange(n_questions+1), br3,'b', label='c-m')
plt.plot(np.arange(n_questions+1), br6,'k--', label='rand')
plt.plot(np.arange(n_questions+1), br7,'r--', label='h')
plt.plot(np.arange(n_questions+1), br8,'b--', label='m')
plt.plot(np.arange(n_questions+1), br9,'y--', label='r')
plt.legend()
#plt.show()
tikz_save('brier.tex')