Sepsis_Label_Extraction/src/visualization/table_functions.py at main · datasig-ac-uk/Sepsis_Label_Extraction · 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
import sys

import numpy as np
import pandas as pd
from sklearn.metrics import roc_auc_score, roc_curve, auc

import constants
import features.mimic3_function as mimic3_myfunc
from visualization.patientlevel_function import output_at_metric_level


headers = ['H1', 'H2', 'H3']


def instance_level_auc_pd_threemodels(labels_list_list, probs_list_list,
                                      models=constants.MODELS, definitions=constants.FEATURES,
                                      pd_save_name=None):
    """
        instance level auc pd outout for all three models
    """
    results = []

    for model in range(len(models)):

        aucs = []

        for defi in range(len(definitions)):
            fpr, tpr, _ = roc_curve(
                labels_list_list[model][defi], probs_list_list[model][defi])
            roc_auc = auc(fpr, tpr)

            aucs.append(roc_auc)

        results.append([models[model], "{:.3f}".format(aucs[0]),
                        "{:.3f}".format(aucs[1]),
                        "{:.3f}".format(aucs[-1])])

    output_df = pd.DataFrame(results, columns=['Model', 'H1', 'H2', 'H3'])

    if pd_save_name is not None:

        output_df.to_csv(pd_save_name + ".csv")

    else:

        return output_df


def patient_level_auc_pd_threemodels(fprs_list_list, tprs_list_list,
                                     models=constants.MODELS, headers=headers,
                                     definitions=constants.FEATURES,
                                     pd_save_name=None,
                                     numerics_format="{:.3f}", for_write=True):
    """
        patient level auc pd output for all three models
    """
    results = []

    for model in range(len(models)):

        aucs = []

        for defi in range(len(definitions)):

            roc_auc = auc(fprs_list_list[model]
                          [defi], tprs_list_list[model][defi])

            if for_write:
                aucs += [numerics_format.format(roc_auc) + ' &']
            else:
                aucs += [numerics_format.format(roc_auc)]

        results.append([models[model]] + aucs)

    output_df = pd.DataFrame(results, columns=['Model'] + headers)

    if pd_save_name is not None:

        output_df.to_csv(pd_save_name + ".csv")

    else:

        return output_df


def patient_level_output_pd_threemodels(some_list_list, metric_seq_list_list,
                                        models=constants.MODELS, headers=headers, definitions=constants.FEATURES,
                                        metric_required=[0.375], numerics_format="{:.2%}",
                                        operator=lambda x: x, for_write=True, pd_save_name=None):
    """
        patient level pd-format output for all three models
    """
    results = []

    for model in range(len(models)):

        outputs_current = []

        for defi in range(len(definitions)):

            output_current = output_at_metric_level(operator(some_list_list[model][defi]),
                                                    metric_seq_list_list[model][defi],
                                                    metric_required=metric_required)
            if for_write:
                outputs_current += [
                    numerics_format.format(output_current) + ' &']
            else:
                outputs_current += [numerics_format.format(output_current)]

        results.append([models[model]] + outputs_current)

    output_df = pd.DataFrame(results, columns=['Model'] + headers)

    if pd_save_name is not None:

        output_df.to_csv(pd_save_name + ".csv")

    else:

        return output_df