-
Notifications
You must be signed in to change notification settings - Fork 0
/
measure_indicators.py
136 lines (116 loc) · 5.47 KB
/
measure_indicators.py
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
import joblib
import numpy as np
import pandas as pd
from common import (
cleanup_files,
convert_dtypes,
force_refresh,
get_logger,
get_path,
import_dataset,
import_features,
initialize,
measured,
open_metadata,
)
initialize()
@convert_dtypes
def add_stale_warning(stales):
def find_stale_warning(events):
events["is_warning"] = events["event"].isin(["commented", "labeled"]) & ~(
events.shift(-1)["event"].eq("closed")
& (events.shift(-1)["time"] - events["time"]).le(np.timedelta64(1, "m"))
)
return events
return stales.groupby("pull_number", group_keys=False).apply(find_stale_warning)
@convert_dtypes
def measure_activity(stales):
def measure_month(events):
return pd.Series(
{
"events": len(events),
"staled": events.index.get_level_values("pull_number").nunique(),
"warned": events.query("is_warning").index.get_level_values("pull_number").nunique(),
"closed": events.query("event == 'closed'").index.get_level_values("pull_number").nunique(),
}
)
activity = stales.groupby("month", group_keys=False).apply(measure_month)
return activity.reindex(range(activity.index.min(), activity.index.max() + 1), fill_value=0)
def export_features_fixed(project, features):
features.reset_index().set_index("project").to_csv(get_path("features_fixed", project))
def export_activity(project, activity):
activity.assign(project=project).reset_index().set_index("project").to_csv(get_path("activity", project))
def export_indicators(project, indicators):
indicators.assign(project=project).set_index("project").to_csv(get_path("indicators", project))
def measure_indicators(project):
logger = get_logger(__file__, modules={"sqlitedict": "WARNING"})
logger.info(f"{project}: Measuring indicators")
dataset = import_dataset(project)
features = import_features(project)
metadata = open_metadata(project)
stales = dataset.query("is_stale").copy()
if project == "automattic/wp-calypso":
first_stale = pd.Timestamp("2019-04-20 05:46:48")
stales = stales.query("time >= @first_stale").copy()
else:
first_stale = stales["time"].min()
last_stale = stales["time"].max()
stales["month"] = (stales["time"] - first_stale) // np.timedelta64(1, "M")
stales = add_stale_warning(stales)
activity = measure_activity(stales)
characteristics = features.select_dtypes(include="number").columns
features = features[features[characteristics].ge(0).all(axis="columns")].copy()
features["opened_month"] = (features["opened_at"] - first_stale) // np.timedelta64(1, "M")
features["resolved_month"] = (features["resolved_at"] - first_stale) // np.timedelta64(1, "M")
monthly = pd.DataFrame()
for month in range(
features["opened_month"].min(), int(features[["opened_month", "resolved_month"]].max().max() + 1)
):
opened = features.query("opened_month == @month")
resolved = features.query("resolved_month == @month")
monthly.loc[month, "opened_pulls"] = len(opened)
monthly.loc[month, "merged_pulls"] = len(resolved.query("is_merged"))
monthly.loc[month, "closed_pulls"] = len(resolved.query("is_closed"))
monthly.loc[month, "active_contributors"] = opened["contributor"].nunique()
monthly["open_pulls"] = (
(monthly["opened_pulls"] - monthly["merged_pulls"] - monthly["closed_pulls"]).cumsum().shift(1, fill_value=0)
)
monthly["workload"] = monthly["opened_pulls"] + monthly["open_pulls"]
monthly.index.name = "month"
activity = monthly.join(activity).fillna(0)
indicators = (
features.groupby(["resolved_month", "is_merged"], as_index=False)
.agg(dict.fromkeys(characteristics, "mean"))
.merge(activity, left_on="resolved_month", right_index=True)
)
indicators["time"] = indicators["resolved_month"] + abs(indicators["resolved_month"].min()) + 1
indicators["adoption"] = indicators["resolved_month"] >= 0
indicators["time_since_adoption"] = indicators["resolved_month"].apply(lambda month: month + 1 if month >= 0 else 0)
indicators["first_stale_time"] = first_stale
indicators["last_stale_time"] = last_stale
indicators["stale_activity_period"] = (last_stale - first_stale) / np.timedelta64(1, "M")
indicators["age_at_adoption"] = (
first_stale - pd.Timestamp(metadata["created_at"]).tz_localize(None)
) / np.timedelta64(1, "M")
features_before = features.query("opened_month < 0")
indicators["pulls_at_adoption"] = len(features_before)
indicators["contributors_at_adoption"] = features_before["contributor"].nunique()
indicators["maintainers_at_adoption"] = dataset.query("time < @first_stale and is_core")["actor"].nunique()
export_features_fixed(project, features)
export_activity(project, activity)
export_indicators(project, indicators)
def main():
projects = []
for project in measured():
if cleanup_files(["features_fixed", "activity", "indicators"], force_refresh(), project):
projects.append(project)
else:
print(f"Skip measuring indicators for project {project}")
with joblib.Parallel(n_jobs=-1, verbose=1) as parallel:
parallel(joblib.delayed(measure_indicators)(project) for project in projects)
if __name__ == "__main__":
try:
main()
except KeyboardInterrupt:
print("Stop measuring indicators")
exit(1)