main.py 9.12 KB
Newer Older
1
2
3
4
from flask import Flask, jsonify, request
import pandas as pd
import numpy as np
from flask_cors import CORS
5
from collections import defaultdict, Counter
6
from time import time
7
8
9
import os.path
import json
from sklearn import preprocessing
10
import orjson
11
import dask.dataframe as dd
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
12
13
14
15
import bigwig
import bbi

reload = False
16
17
18
19
20
21
22
23
24
25

app = Flask(__name__)
CORS(app)

@app.route('/', methods=['GET'])
def index():
    return "hi"

@app.route('/read-data', methods=['GET'])
def read_data():
26
    t0 = time()
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
27
28
29
30
    size = bbi.chromsizes('test.bigWig')['chr1']
    bins = 100000
    data = bigwig.get('test.bigWig', 'chr1', 0, size, bins)
    print(data.shape)
31
    response = {
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
32
33
        "index": list(range(0, size, int(size/(bins)))),
        "values": data.tolist()
34
    }
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
35
    response = orjson.dumps(response)
36
    print('Data read: ' + str(time()-t0))
37
38
39
40
    return response

@app.route('/create-windows', methods=['POST'])
def create_windows():
41
    t0 = time()
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
42
    if reload:
43
44
        raw_data = request.json
        window_size = int(raw_data['parameters']["windowsize"])
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
45
46
47
        data = bigwig.chunk(
            'test.bigWig',
            12000,
48
49
            int(12000 / window_size),
            int(12000 / 6),
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
50
51
52
53
            ['chr1'],
            verbose=True,
        )
        print(data.shape)
54
        np.save('processed-data', data)
55
    print('Windows created: ' + str(time()-t0))
56
    return '1'
57
58
59

@app.route('/create-tables', methods=['POST'])
def create_tables():
60
61
62
63
64
65
    t0 = time()
    data = np.load('processed-data.npy')
    raw_data = orjson.loads(request.data)
    window_size = int(raw_data['parameters']["windowsize"])
    hash_size = int(raw_data['parameters']["hashsize"])
    table_size = int(raw_data['parameters']["tablesize"])
66

67
    print('Starting: ' + str(time()-t0))
68
    tables_hash_function = [np.random.uniform(-100, 100, size=(window_size, hash_size)) for _ in range(table_size)]
69
70
71
72
73
    print('Init time: ' + str(time() - t0))
    tables = []
    for index in range(table_size):
        t1 = time()
        table = defaultdict(list)
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
74
        signatures_bool = np.dot(data, tables_hash_function[index]) > 0
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
        signatures = [''.join(['1' if x else '0' for x in lst]) for lst in signatures_bool]
        for i in range(len(signatures)):
            table[signatures[i]].append(i)
        print(time()-t1)
        tables.append(table)

    print('Creation time: ' + str(time() - t0))
    hash_functions = np.array(tables_hash_function).tolist()
    response = {}
    for table_index in range(table_size):
        response[str(table_index)] = {
            "hash": hash_functions[table_index],
            "entries": tables[table_index]
        }
    response = orjson.dumps(response)
    return response

92
93
@app.route('/query', methods=['POST'])
def query():
94
    t0 = time()
95
    raw_data = orjson.loads(request.data)
96
    window = raw_data['window']
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
    if isinstance(window, int):
        output = np.load('processed-data.npy')[window]
        response = orjson.dumps(output.tolist())
        print("Query done: " + str(time() - t0))
        return response
    else :
        output = preprocessing.minmax_scale(window, (-1, 1))
        response = orjson.dumps(output.tolist())
        print("Query done: " + str(time()-t0))
        return response

@app.route('/window', methods=['POST'])
def window():
    t0 = time()
    raw_data = orjson.loads(request.data)
    indices = raw_data['indices']
    output = np.load('processed-data.npy')[indices]
114
    response = orjson.dumps(output.tolist())
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
115
    print("Query done: " + str(time() - t0))
116
117
118
119
120
121
122
    return response

@app.route('/similarity', methods=['POST'])
def similarity():
    t0 = time()
    raw_data = orjson.loads(request.data)
    window = raw_data['query']
123
124
125
    tables = raw_data["tables"]
    neighbours = []

Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
126
    output = defaultdict(list)
127
128

    for t in tables.values():
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
129
        signature_bool = np.dot(window, t["hash"]) > 0
130
        signature = ''.join(['1' if x else '0' for x in signature_bool])
131
        neighbours.extend(t["entries"][signature])
132
133
    neighbours_with_frequency = dict(Counter(neighbours))
    for index, frequency in neighbours_with_frequency.items():
134
        output[str(frequency)].append(index)
135
    response = orjson.dumps(output)
136
137
138
139
140
141
142
143
144
145
146
    print("Similarity done: " + str(time()-t0))
    return response

@app.route('/average-progress', methods=['POST'])
def average_progress():
    t0 = time()
    raw_data = orjson.loads(request.data)
    all_windows = raw_data['windows']
    data = np.load('processed-data.npy')
    output = []
    actual_windows = []
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
147
    print("Starting average progress")
148
149
150
    print("Initialized: " + str(time() - t0))
    for windows in all_windows:
        t1 = time()
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
151
        actual_windows.extend(data[windows])
152
153
154
        if len(actual_windows) == 0:
            output.append([])
            continue
155
156
157
        max_values = np.maximum.reduce(actual_windows).tolist()
        min_values = np.minimum.reduce(actual_windows).tolist()
        average_values = (np.sum(actual_windows, 0)/len(actual_windows)).tolist()
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
158
        output = [({
159
160
161
            'average': average_values,
            'max': max_values,
            'min': min_values
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
162
        })] + output
163
164
165
        print("Average calculated: " + str(time() - t1))
    response = orjson.dumps(output)
    print("Averages calculated: " + str(time() - t0))
166
167
    return response

168
169
@app.route('/average-table', methods=['POST'])
def average_table():
170
171
172
173
174
    t0 = time()
    raw_data = orjson.loads(request.data)
    all_windows = raw_data['windows']
    data = np.load('processed-data.npy')
    output = []
175
    print("Initialized: " + str(time() - t0))
176
177
    for windows in all_windows:
        t1 = time()
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
178
        actual_windows = data[windows]
179
        print(len(actual_windows))
180
181
182
183
184
185
186
        average_values = np.average(actual_windows, 0)
        # average_values = (np.sum(actual_windows, 0) / len(actual_windows))
        std_values = np.std(actual_windows, 0)
        max_values = average_values + std_values
        min_values = average_values - std_values
        # max_values = np.maximum.reduce(actual_windows).tolist()
        # min_values = np.minimum.reduce(actual_windows).tolist()
187
        output.append({
188
189
190
            'average': average_values.tolist(),
            'max': max_values.tolist(),
            'min': min_values.tolist()
191
        })
192
193
        print("Average calculated: " + str(time() - t1))
    response = orjson.dumps(output)
194
    print("Averages calculated: " + str(time() - t0))
195
196
197
198
199
    return response

@app.route('/update', methods=['POST'])
def update():
    t0 = time()
200
    print("Start")
201
    raw_data = orjson.loads(request.data)
202
203
    print("Data loaded: " + str(time() - t0))
    data = np.load('processed-data.npy')
204
205
    label_data = raw_data["labelData"]
    tables = raw_data["tables"]
206
    window = raw_data["query"]
207
208
209
210
211
212

    window_size = int(raw_data['parameters']["windowsize"])
    hash_size = int(raw_data['parameters']["hashsize"])
    table_size = int(raw_data['parameters']["tablesize"])
    new_tables = []

213
214
    correct_indices = [int(index) for index, value in label_data.items() if value is True]
    incorrect_indices = [int(index) for index, value in label_data.items() if value is False]
215

216
    print("Initialized: " + str(time() - t0))
217
218
    for t in tables.values():
        valid = True
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
219
        signature = ''.join((np.dot(window, t["hash"]) > 0).astype('int').astype('str'))
220
        neighbours = t["entries"][signature]
221
222
223
224
225
226
227
228
229
230
        for index in correct_indices:
            if index not in neighbours:
                valid = False
                break
        for index in incorrect_indices:
            if index in neighbours:
                valid = False
                break
        if valid:
            new_tables.append(t)
231
232
233
234
235
236
    print("Filtered good tables: " + str(time() - t0))
    for index in range(table_size - len(new_tables)):
        entries = defaultdict(list)
        t1 = time()
        while True:
            hash_function = np.random.randn(window_size, hash_size)
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
237
            correct_signatures = [''.join((np.dot(data[i], hash_function) > 0).astype('int').astype('str')) for
238
                                  i in
239
                                  correct_indices]
240
241
            incorrect_signatures = [''.join((np.dot(data[i], hash_function) > 0).astype('int').astype('str')) for
                                    i
242
243
244
245
246
247
248
                                    in incorrect_indices]
            if correct_signatures.count(correct_signatures[0]) == len(
                    correct_signatures) and incorrect_signatures.count(
                    correct_signatures[0]) == 0:
                break
        print("first: " + str(time() - t1))
        t2 = time()
Kruyff,D.L.W. (Dylan)'s avatar
Kruyff,D.L.W. (Dylan) committed
249
        signatures_bool = np.dot(data, hash_function) > 0
250
251
252
253
254
255
256
257
        signatures = [''.join(['1' if x else '0' for x in lst]) for lst in signatures_bool]
        for i in range(len(signatures)):
            entries[signatures[i]].append(i)
        print("second: " + str(time() - t2))
        new_tables.append({
            "hash": hash_function.tolist(),
            "entries": entries
        })
258

259
260
261
262
263
264
265
266
    print('Update time: ' + str(time() - t0))
    response = {}
    for table_index in range(len(new_tables)):
        response[table_index] = {
            "hash": new_tables[table_index]["hash"],
            "entries": new_tables[table_index]["entries"]
        }
    response = jsonify(response)
267
    return response