Spatio-Temporal-Classifier/utils.py at main · Ruhallah93/Spatio-Temporal-Classifier · 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
import os
import csv
import numpy as np
import pandas as pd
from tensorflow.keras.losses import MeanSquaredError
from sklearn.metrics import accuracy_score, recall_score, f1_score, precision_score


def get_segments_a_decision_window(segment_size, segment_overlap, decision_size):
    segment_overlap_size = segment_size * segment_overlap
    return int((decision_size - segment_size) / (segment_size - segment_overlap_size) + 1)


def AveragingProbabilities(y_truth, y_prediction, s, r):
    df = []
    y_dw_truth = []
    c = y_prediction.shape[1]
    r = int(np.floor(s * r))
    for _id in np.unique(y_truth):
        subset = y_prediction[np.where(y_truth == _id)]
        n = subset.shape[0]
        o = int(np.floor((n - r) / (s - r)))
        for i in range(o):
            row = np.zeros(c)
            for j in range(s):
                row += subset[(i * s) + j]
            df.append(row / s)
            y_dw_truth.append(_id)
    df = np.array(df)
    y_dw_truth = np.asarray(pd.get_dummies(y_dw_truth), dtype=np.int8)
    return df, y_dw_truth


def MajorityVote(y_truth, y_prediction, s, r):
    df = []
    y_dw_truth = []
    c = y_prediction.shape[1]
    r = int(np.floor(s * r))
    # Make prior prediction to one-hot
    y_categorical_pred = np.zeros_like(y_prediction)
    y_categorical_pred[np.arange(len(y_prediction)), y_prediction.argmax(1)] = 1

    for _id in np.unique(y_truth):
        subset = y_categorical_pred[np.where(y_truth == _id)]
        n = subset.shape[0]
        o = int(np.floor((n - r) / (s - r)))
        for i in range(o):
            row = np.zeros(c)
            for j in range(s):
                row += subset[(i * s) + j]
            df.append(row / s)
            y_dw_truth.append(_id)
    df = np.array(df)
    y_dw_truth = np.asarray(pd.get_dummies(y_dw_truth), dtype=np.int8)
    return df, y_dw_truth


# p = MaximumScore(np.array([1, 1, 1, 1, 2, 2, 2, 2]), np.array(
#     [[0.1, 0.8], [1.1, 0.2], [0.1, 0.5], [0.1, 0.7], [0.8, 0.2], [0.9, 0.2], [0.3, 0.6], [0.2, 0.8]]), 2, 0.5)
# print(p)
# p = MajorityVote(np.array([1, 1, 1, 1, 2, 2, 2, 2]), np.array(
#     [[0.1, 0.8], [1.1, 0.2], [0.1, 0.5], [0.1, 0.7], [0.8, 0.2], [0.9, 0.2], [0.3, 0.6], [0.2, 0.8]]), 2, 0.5)
# print(p)

def analysis_model(y_pred, y_real_raw, segment_size, segment_overlap, decision_size, decision_overlap):
    result = {'Core': {}, 'MV': {}, 'MS': {}}
    loss_fn = MeanSquaredError()

    result['Core']['mse_loss'] = loss_fn(np.asarray(pd.get_dummies(y_real_raw), dtype=np.int8),
                                         np.asarray(y_pred, dtype=np.float)).numpy()
    y_pred_arg = np.argmax(y_pred, axis=1)
    result['Core']['accuracy'] = accuracy_score(y_real_raw, y_pred_arg)
    result['Core']['precision'] = precision_score(y_real_raw, y_pred_arg, average='macro')
    result['Core']['recall'] = recall_score(y_real_raw, y_pred_arg, average='macro')
    result['Core']['f1'] = f1_score(y_real_raw, y_pred_arg, average='macro')

    segments_a_decision_window = get_segments_a_decision_window(segment_size, segment_overlap, decision_size)

    # Maximum Score
    y_pred_labels, y_real = AveragingProbabilities(y_truth=y_real_raw, y_prediction=y_pred,
                                                   s=segments_a_decision_window,
                                                   r=decision_overlap)

    result['MS']['mse_loss'] = loss_fn(y_real, y_pred_labels).numpy()
    temp = y_pred_labels.copy()
    y_pred_labels = np.zeros_like(temp)
    y_pred_labels[np.arange(len(temp)), temp.argmax(1)] = 1
    result['MS']['accuracy'] = accuracy_score(y_real, y_pred_labels)
    result['MS']['precision'] = precision_score(y_real, y_pred_labels, average='macro')
    result['MS']['recall'] = recall_score(y_real, y_pred_labels, average='macro')
    result['MS']['f1'] = f1_score(y_real, y_pred_labels, average='macro')

    # Majority Voting
    y_pred_labels, y_real = MajorityVote(y_truth=y_real_raw, y_prediction=y_pred,
                                         s=segments_a_decision_window,
                                         r=decision_overlap)

    result['MV']['mse_loss'] = loss_fn(y_real, y_pred_labels).numpy()
    temp = y_pred_labels.copy()
    y_pred_labels = np.zeros_like(temp)
    y_pred_labels[np.arange(len(temp)), temp.argmax(1)] = 1
    result['MV']['accuracy'] = accuracy_score(y_real, y_pred_labels)
    result['MV']['precision'] = precision_score(y_real, y_pred_labels, average='macro')
    result['MV']['recall'] = recall_score(y_real, y_pred_labels, average='macro')
    result['MV']['f1'] = f1_score(y_real, y_pred_labels, average='macro')

    return result


def save_result(log_dir, data: dict):
    if not os.path.exists(log_dir):
        os.makedirs(log_dir)

    # Save to file
    with open(log_dir + 'statistics.txt', 'a') as f:
        f.write('\n==========***==========\n')
        f.write(str(data))
        f.write('\n')

    csv_file = log_dir + 'statistics.csv'
    file_exists = os.path.isfile(csv_file)
    try:
        with open(csv_file, 'a') as csvfile:
            writer = csv.writer(csvfile)
            if not file_exists:
                writer.writerow(data.keys())
            writer.writerow(data.values())
            csvfile.close()
    except IOError:
        print("I/O error")


def cross_entropy3(p, q, ets=1e-15):
    a = [p[i] * np.log(q[i] + ets) for i in range(len(p))]
    return (-sum(a)) + np.std(np.array(a)[np.where(np.array(q) != 1)])