-
Notifications
You must be signed in to change notification settings - Fork 0
/
server.py
65 lines (48 loc) · 1.69 KB
/
server.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
from training import build_model, fit_transform
import numpy as np
from keras.optimizers import Adam
from keras.models import load_model
from random import uniform, seed
from flask import Flask, request, jsonify
from flask_cors import CORS
app = Flask(__name__)
CORS(app)
@app.route('/')
def hello_world():
return 'Hello World!'
@app.route('/post', methods=['POST'])
def post():
data = request.get_json()
print(data) # dictionary
N = len(data["ids"])
# print(N)
times = np.array([[data["month"], data["weekday"], data["hour"]] for _ in range(N)])
ids = np.array([[e] for e in data["ids"]])
refined_ids = fit_transform(ids.reshape(-1, 1), max_val, min_val)
pred = model.predict([times, refined_ids]) # month, weekday, hour | id
print(pred)
pred = pred.tolist()
res = [e[0] for e in pred]
if data["adj"] != 0:
# res = [e - 2 for e in res] # Adjusting
loss = 2
seed(data["month"] * 1000 + data["weekday"] * 100 + data["hour"])
res = [e + uniform(-loss, loss) for e in res] # Adjusting loss
else:
pass
res = [round(e) if e > 0 else round(e) - 1 for e in res] # Adjusting minus values
print(res)
return jsonify({"res": res})
if __name__ == '__main__':
# print("DEV")
"""model"""
input_shape_time = (3, )
input_shape_id = (1, )
input_shapes = [input_shape_time, input_shape_id]
output_shape = 1
model = build_model(input_shapes, output_shape)
model.compile(loss='mae', optimizer=Adam(), metrics=['mse', 'acc'])
model = load_model('citibike_DNN_model.h5')
"""inference"""
max_val, min_val = 3911, 72 # max: 3911 min: 72
app.run(host='0.0.0.0', port=8327)