analyse_test/calibration_processor.py at main · himyss/analyse_test · 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
import json
from calibration_data import CalibrationData

class CalibrationProcessor:
    @staticmethod
    def load_time_ranges(json_path: str, user_id: int) -> dict:
        """
        Загрузить тайминги фаз из JSON в формате:
        {
          "704": [
            { "shelf": "Baseline",  "time_range": [0.0, 30.0] },
            { "shelf": "Positive",  "time_range": [30.0, 60.0] },
            { "shelf": "Neutral_2",   "time_range": [60.0, 90.0] },
            { "shelf": "Negative",  "time_range": [90.0, 120.0] }
          ],
          ...
        }
        Возвращает dict вида {"Baseline": [...], "Positive": [...], ...}
        """
        with open(json_path, 'r', encoding='utf-8') as f:
            all_timings = json.load(f)

        if str(user_id) not in all_timings:
            raise ValueError(f"No timings for user {user_id}")
        return {
            item['shelf']: item['time_range']
            for item in all_timings[str(user_id)]
        }

    @staticmethod
    def compute_baseline(filtered_data, time_ranges: dict) -> tuple[float, float]:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['Baseline']
        mask = (
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1)
        )
        segment = df.loc[mask, 'avg_pupildiameter_smooth']
        if segment.empty:
            raise ValueError(f"No valid data in baseline [{t0}, {t1}]")
        return segment.mean(), segment.std(ddof=1)

    @staticmethod
    def compute_positive_rise_speed(filtered_data, time_ranges: dict, baseline_mean: float) -> float:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['Positive']
        seg = df.loc[
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1),
            ['relative_timestamp', 'avg_pupildiameter_smooth']
        ]
        if seg.empty:
            raise ValueError(f"No valid data in positive [{t0}, {t1}]")
        # момент пика
        idx_peak = seg['avg_pupildiameter_smooth'].idxmax()
        t_peak   = seg.loc[idx_peak, 'relative_timestamp']
        val_peak = seg.loc[idx_peak, 'avg_pupildiameter_smooth']
        return (val_peak - baseline_mean) / (t_peak - t0)

    @staticmethod
    def compute_positive_fall_speed(filtered_data, time_ranges: dict, baseline_mean: float) -> float:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['Positive']
        seg = df.loc[
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1),
            ['relative_timestamp', 'avg_pupildiameter_smooth']
        ]
        if seg.empty:
            raise ValueError(f"No valid data in positive [{t0}, {t1}]")
        idx_peak = seg['avg_pupildiameter_smooth'].idxmax()
        t_peak   = seg.loc[idx_peak, 'relative_timestamp']
        val_peak = seg.loc[idx_peak, 'avg_pupildiameter_smooth']
        # значение в конце фазы (берём последнее по времени)
        idx_end  = seg['relative_timestamp'].idxmax()
        t_end    = t1
        val_end  = seg.loc[idx_end, 'avg_pupildiameter_smooth']
        return (val_peak - val_end) / (t_end - t_peak)

    @staticmethod
    def compute_negative_rise_speed(filtered_data, time_ranges: dict, baseline_mean: float) -> float:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['negative']
        seg = df.loc[
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1),
            ['relative_timestamp', 'avg_pupildiameter_smooth']
        ]
        if seg.empty:
            raise ValueError(f"No valid data in negative [{t0}, {t1}]")
        idx_peak = seg['avg_pupildiameter_smooth'].idxmax()
        t_peak   = seg.loc[idx_peak, 'relative_timestamp']
        val_peak = seg.loc[idx_peak, 'avg_pupildiameter_smooth']
        return (val_peak - baseline_mean) / (t_peak - t0)

    @staticmethod
    def compute_negative_fall_speed(filtered_data, time_ranges: dict, baseline_mean: float) -> float:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['Negative']
        seg = df.loc[
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1),
            ['relative_timestamp', 'avg_pupildiameter_smooth']
        ]
        if seg.empty:
            raise ValueError(f"No valid data in negative [{t0}, {t1}]")
        idx_peak = seg['avg_pupildiameter_smooth'].idxmax()
        t_peak   = seg.loc[idx_peak, 'relative_timestamp']
        val_peak = seg.loc[idx_peak, 'avg_pupildiameter_smooth']
        idx_end  = seg['relative_timestamp'].idxmax()
        t_end    = t1
        val_end  = seg.loc[idx_end, 'avg_pupildiameter_smooth']
        return (val_peak - val_end) / (t_end - t_peak)

    @staticmethod
    def compute_neutral2_recovery_speed(filtered_data, time_ranges: dict, baseline_mean: float) -> float:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['Neutral_2']
        seg = df.loc[
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1),
            ['relative_timestamp', 'avg_pupildiameter_smooth']
        ]
        if seg.empty:
            raise ValueError(f"No valid data in neutral2 [{t0}, {t1}]")
        idx_start = seg['relative_timestamp'].idxmin()
        idx_end   = seg['relative_timestamp'].idxmax()
        val_start = seg.loc[idx_start, 'avg_pupildiameter_smooth']
        val_end   = seg.loc[idx_end,   'avg_pupildiameter_smooth']
        return (val_start - val_end) / (t1 - t0)

    @staticmethod
    def compute_positive_mean_ratio(filtered_data, time_ranges: dict, baseline_mean: float) -> float:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['Positive']
        seg = df.loc[
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1),
            'avg_pupildiameter_smooth'
        ]
        if seg.empty:
            raise ValueError(f"No valid data in positive [{t0}, {t1}]")
        return seg.mean() / baseline_mean

    @staticmethod
    def compute_negative_mean_ratio(filtered_data, time_ranges: dict, baseline_mean: float) -> float:
        df = filtered_data.get_data()
        t0, t1 = time_ranges['Positive']
        seg = df.loc[
            (df['is_valid']) &
            (df['relative_timestamp'] >= t0) &
            (df['relative_timestamp'] <= t1),
            'avg_pupildiameter_smooth'
        ]
        if seg.empty:
            raise ValueError(f"No valid data in negative [{t0}, {t1}]")
        return seg.mean() / baseline_mean

    @staticmethod
    def process(filtered_data, user_id: int, json_path: str) -> CalibrationData:
        time_ranges = CalibrationProcessor.load_time_ranges(json_path, user_id)
        baseline_mean, baseline_std = CalibrationProcessor.compute_baseline(filtered_data, time_ranges)
        pos_rise = CalibrationProcessor.compute_positive_rise_speed(filtered_data, time_ranges, baseline_mean)
        pos_fall = CalibrationProcessor.compute_positive_fall_speed(filtered_data, time_ranges, baseline_mean)
        neg_rise = CalibrationProcessor.compute_negative_rise_speed(filtered_data, time_ranges, baseline_mean)
        neg_fall = CalibrationProcessor.compute_negative_fall_speed(filtered_data, time_ranges, baseline_mean)
        neutral2_r = CalibrationProcessor.compute_neutral2_recovery_speed(filtered_data, time_ranges, baseline_mean)
        pos_ratio = CalibrationProcessor.compute_positive_mean_ratio(filtered_data, time_ranges, baseline_mean)
        neg_ratio = CalibrationProcessor.compute_negative_mean_ratio(filtered_data, time_ranges, baseline_mean)

        return CalibrationData(
            user_id=user_id,
            baseline_mean=baseline_mean,
            baseline_std=baseline_std,
            positive_rise_speed=pos_rise,
            negative_rise_speed=neg_rise,
            positive_fall_speed=pos_fall,
            negative_fall_speed=neg_fall,
            neutral2_recovery_speed=neutral2_r,
            positive_mean_ratio=pos_ratio,
            negative_mean_ratio=neg_ratio
        )