clustever/api.py at master · cortext/clustever · 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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# by Antoine Mazières (http://mazier.es ; {github|twitter}@mazieres)
# Cortext Lab -- http://www.cortext.net/
#
from sklearn import cluster, decomposition
from sklearn.datasets import load_iris, load_boston
from inspect import getargspec, ismethod
from itertools import chain
import pandas as pd
from random import shuffle
import json

from flask import Flask, request, Response
app = Flask(__name__)

import unittest
import requests

class TestClustever(unittest.TestCase):

    def test_query(self):
        q = {"clustering": {"KMeans": {"n_clusters": 3}}, "dimred": {"PCA": {"n_components": 2}}, "dataset": "iris"}
        r = requests.get('http://{}:{}'.format(addr, port), params={'q': json.dumps(q)})
        tested = r.status_code
        expected = 200
        msg = '\nExpected:\n{}\nGot:\n{}'.format(expected, tested)
        self.assertEqual(expected, tested, msg=msg)

    def test_res(self):
        q = {"clustering": {"KMeans": {"n_clusters": 3}}, "dimred": {"PCA": {"n_components": 2}}, "dataset": "iris"}
        r = requests.get('http://{}:{}'.format(addr, port), params={'q': json.dumps(q)})
        self.assertEqual(3, len(r.json()))
        self.assertEqual(150, sum([len(x['data']) for x in r.json()]))
        self.assertEqual(2, len(r.json()[0]['data'][0]))


# Set to 0.0.0.0 for www (NOT SAFE !)
addr = '127.0.0.1'
port = 5001

# Maps all functions in cluster and decomposition
# to their names.
fns = dict(chain(*[
        {x:f.__dict__[x] for x in f.__all__}.items()
        for f in [cluster, decomposition]
    ]))

# Maps all functions arguments and defaults
# to their function names.
args = {name:dict(zip(args.args[1:], list(args.defaults)))
        for name, args in {k:getargspec(v.__init__)
        for k, v in fns.iteritems()
        if ismethod(v.__init__)
        and getargspec(v.__init__).defaults != None}.iteritems()}

fns = {k:v for k, v in fns.iteritems() if k in args}

# Datasets
dss = {ds.DESCR.split(' ')[0].lower():{
            'title': ds.DESCR.split('\n')[0],
            'df': pd.DataFrame(ds.data)}
        for ds in [load_iris(), load_boston()]}

# Colors
colors = [ # http://www.december.com/html/spec/colorrgbadec.html
    'rgba(0, 0, 0, 0.7)', # black
    'rgba(0, 0, 128, 0.7)', # navy
    'rgba(0, 0, 255, 0.7)', # blue
    'rgba(0, 128, 0, 0.7)', # green
    'rgba(0, 128, 128, 0.7)', # teal
    'rgba(0, 255, 0, 0.7)', # lime
    'rgba(0, 255, 255, 0.7)', # aqua
    'rgba(128, 0, 0, 0.7)', # marroon
    'rgba(128, 0, 128, 0.7)', # purple
    'rgba(128, 128, 0, 0.7)', # olive
    'rgba(128, 128, 128, 0.7)', # gray
    'rgba(192, 192, 192, 0.7)', # silver
    'rgba(255, 0, 0, 0.7)', # red
    'rgba(255, 0, 255, 0.7)', # fuchsia
    'rgba(255, 255, 0, 0.7)', # yellow
]
shuffle(colors)


class Clustever(object):
    def __init__(self, q):
        self.q = q
        self.df = dss[self.q['dataset']]['df']
        self.df_norm = self.norm_ds()
        self.clusters = self.clustering()
        self.twod = self.dimred()

    def clustering(self):
        fn_name, fn_args = self.q['clustering'].items()[0]
        fn = fns[fn_name]
        return fn(**fn_args).fit_predict(self.df_norm.values)

    def dimred(self):
        fn_name, fn_args = self.q['dimred'].items()[0]
        fn = fns[fn_name]
        return fn(**fn_args).fit_transform(self.df_norm.values)

    def norm_ds(self):
        return (self.df - self.df.mean()) / self.df.std()

    def series(self):
        self.series = []
        i = 0
        for g in set(self.clusters):
            clust = {
                'name': 'Cluster {}'.format(i),
                'color': colors[i],
                'data': self.twod[self.clusters == g].tolist()
            }
            self.series.append(clust)
            i += 1
        return self.series


@app.route("/")
def hello():
    req = request.args
    print req
    if 'q' in req:
        callback = req['callback']
        q = json.loads(req['q'])
        c = Clustever(q)
        res = c.series()
        return Response("{}({})".format(callback, json.dumps(res)), mimetype='application/javascript')
    else:
        print "toto"
        return "None"

if __name__ == "__main__":
    app.run(host=addr, port=port, debug=True)