File size: 7,221 Bytes
d67d34a
 
 
 
 
 
c431e93
d67d34a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c431e93
 
 
d67d34a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
195
196
197
198
199
200
201
202
203
204
205
import os
from json import dumps
from typing import Dict, List, Union

import numpy as np
import pandas as pd
from fastapi import FastAPI
from gamma.utils import association
from kafka import KafkaProducer
from pydantic import BaseModel
from pyproj import Proj

# Kafak producer
use_kafka = False

try:
    print("Connecting to k8s kafka")
    BROKER_URL = "quakeflow-kafka-headless:9092"
    # BROKER_URL = "34.83.137.139:9094"
    producer = KafkaProducer(
        bootstrap_servers=[BROKER_URL],
        key_serializer=lambda x: dumps(x).encode("utf-8"),
        value_serializer=lambda x: dumps(x).encode("utf-8"),
    )
    use_kafka = True
    print("k8s kafka connection success!")
except BaseException:
    print("k8s Kafka connection error")
    try:
        print("Connecting to local kafka")
        producer = KafkaProducer(
            bootstrap_servers=["localhost:9092"],
            key_serializer=lambda x: dumps(x).encode("utf-8"),
            value_serializer=lambda x: dumps(x).encode("utf-8"),
        )
        use_kafka = True
        print("local kafka connection success!")
    except BaseException:
        print("local Kafka connection error")
print(f"Kafka status: {use_kafka}")

app = FastAPI()

PROJECT_ROOT = os.path.realpath(os.path.join(os.path.dirname(__file__), ".."))
STATION_CSV = os.path.join(PROJECT_ROOT, "tests/stations_hawaii.csv")
# STATION_CSV = os.path.join(PROJECT_ROOT, "tests/stations.csv")  ## ridgecrest


def default_config(config):
    if "degree2km" not in config:
        config["degree2km"] = 111.195
    if "use_amplitude" not in config:
        config["use_amplitude"] = True
    if "use_dbscan" not in config:
        config["use_dbscan"] = True
    if "dbscan_eps" not in config:
        config["dbscan_eps"] = 30.0
    if "dbscan_min_samples" not in config:
        config["dbscan_min_samples"] = 3
    if "method" not in config:
        config["method"] = "BGMM"
    if "oversample_factor" not in config:
        config["oversample_factor"] = 5
    if "min_picks_per_eq" not in config:
        config["min_picks_per_eq"] = 10
    if "max_sigma11" not in config:
        config["max_sigma11"] = 2.0
    if "max_sigma22" not in config:
        config["max_sigma22"] = 1.0
    if "max_sigma12" not in config:
        config["max_sigma12"] = 1.0
    if "dims" not in config:
        config["dims"] = ["x(km)", "y(km)", "z(km)"]
    return config


## set config
config = {"xlim_degree": [-156.32, -154.32], "ylim_degree": [18.39, 20.39], "z(km)": [0, 41]}  ## hawaii
# config = {'xlim_degree': [-118.004, -117.004], 'ylim_degree': [35.205, 36.205], "z(km)": [0, 41]}  ## ridgecrest

config = default_config(config)
config["center"] = [np.mean(config["xlim_degree"]), np.mean(config["ylim_degree"])]
config["x(km)"] = (np.array(config["xlim_degree"]) - config["center"][0]) * config["degree2km"]
config["y(km)"] = (np.array(config["ylim_degree"]) - config["center"][1]) * config["degree2km"]
config["bfgs_bounds"] = [list(config[x]) for x in config["dims"]] + [[None, None]]

for k, v in config.items():
    print(f"{k}: {v}")

## read stations
stations = pd.read_csv(STATION_CSV, delimiter="\t")
stations = stations.rename(columns={"station": "id"})
stations["x(km)"] = stations["longitude"].apply(lambda x: (x - config["center"][0]) * config["degree2km"])
stations["y(km)"] = stations["latitude"].apply(lambda x: (x - config["center"][1]) * config["degree2km"])
stations["z(km)"] = stations["elevation(m)"].apply(lambda x: -x / 1e3)

print(stations)


class Data(BaseModel):
    picks: List[Dict[str, Union[float, str]]]
    stations: List[Dict[str, Union[float, str]]]
    config: Dict[str, Union[List[float], List[int], List[str], float, int, str]]


class Pick(BaseModel):
    picks: List[Dict[str, Union[float, str]]]


def run_gamma(picks, config, stations):

    proj = Proj(f"+proj=sterea +lon_0={config['center'][0]} +lat_0={config['center'][1]} +units=km")

    stations[["x(km)", "y(km)"]] = stations.apply(
        lambda x: pd.Series(proj(longitude=x.longitude, latitude=x.latitude)), axis=1
    )
    stations["z(km)"] = stations["elevation(m)"].apply(lambda x: -x / 1e3)

    catalogs, assignments = association(picks, stations, config, 0, config["method"])

    catalogs = pd.DataFrame(
        catalogs,
        columns=["time"]
        + config["dims"]
        + ["magnitude", "sigma_time", "sigma_amp", "cov_time_amp", "event_index", "gamma_score"],
    )
    catalogs[["longitude", "latitude"]] = catalogs.apply(
        lambda x: pd.Series(proj(longitude=x["x(km)"], latitude=x["y(km)"], inverse=True)), axis=1
    )
    catalogs["depth(m)"] = catalogs["z(km)"].apply(lambda x: x * 1e3)

    assignments = pd.DataFrame(assignments, columns=["pick_index", "event_index", "gamma_score"])
    picks_gamma = picks.join(assignments.set_index("pick_index")).fillna(-1).astype({"event_index": int})

    return catalogs, picks_gamma


@app.post("/predict_stream")
def predict(data: Pick):

    picks = pd.DataFrame(data.picks)
    if len(picks) == 0:
        return {"catalog": [], "picks": []}

    catalogs, picks_gamma = run_gamma(data, config, stations)

    if use_kafka:
        print("Push events to kafka...")
        for event in catalogs.to_dict(orient="records"):
            producer.send("gmma_events", key=event["time"], value=event)

    return {"catalog": catalogs.to_dict(orient="records"), "picks": picks_gamma.to_dict(orient="records")}


@app.post("/predict")
def predict(data: Data):

    picks = pd.DataFrame(data.picks)
    if len(picks) == 0:
        return {"catalog": [], "picks": []}

    stations = pd.DataFrame(data.stations)
    if len(stations) == 0:
        return {"catalog": [], "picks": []}

    assert "latitude" in stations
    assert "longitude" in stations
    assert "elevation(m)" in stations

    config = data.config
    config = default_config(config)

    if "xlim_degree" not in config:
        config["xlim_degree"] = (stations["longitude"].min(), stations["longitude"].max())
    if "ylim_degree" not in config:
        config["ylim_degree"] = (stations["latitude"].min(), stations["latitude"].max())
    if "center" not in config:
        config["center"] = [np.mean(config["xlim_degree"]), np.mean(config["ylim_degree"])]
    if "x(km)" not in config:
        config["x(km)"] = (
            (np.array(config["xlim_degree"]) - config["center"][0])
            * config["degree2km"]
            * np.cos(np.deg2rad(config["center"][1]))
        )
    if "y(km)" not in config:
        config["y(km)"] = (np.array(config["ylim_degree"]) - config["center"][1]) * config["degree2km"]
    if "z(km)" not in config:
        config["z(km)"] = (0, 41)
    if "bfgs_bounds" not in config:
        config["bfgs_bounds"] = [list(config[x]) for x in config["dims"]] + [[None, None]]

    catalogs, picks_gamma = run_gamma(picks, config, stations)

    if use_kafka:
        print("Push events to kafka...")
        for event in catalogs.to_dict(orient="records"):
            producer.send("gamma_events", key=event["time"], value=event)

    return {"catalog": catalogs.to_dict(orient="records"), "picks": picks_gamma.to_dict(orient="records")}


@app.get("/healthz")
def healthz():
    return {"status": "ok"}