Distributed-Reverse-Image-Search/imagesearch_workers.py at master · sahibsin/Distributed-Reverse-Image-Search · 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
138
139

# source activate tensorflow_p27

#run this code with "celery -A imagesearch_workers worker --loglevel=info" on the worker machines

#Then this machine will become a worker, and will be able to run the app task,
#i.e. the imagesearch_tasks or upload_data functions, whenever the server requests it.

import celery
import numpy as np
from copy import deepcopy
import json
from pyflann import * # I use pyflann data structure. You any that you like
import time
from scipy.spatial import distance
import time

# change this to your load balancers!!!
app = celery.Celery('imagesearch_workers.py',
                        broker='amqp://myguest:myguestpwd@ImageBalance-dca102eb6a30882f.elb.us-east-2.amazonaws.com',
                        backend='amqp://myguest:myguestpwd@ImageBalance-dca102eb6a30882f.elb.us-east-2.amazonaws.com')

#Global variables
#Use any global variables that you will need

flann_kdtree = FLANN()
flann_linear = FLANN()
features_train=[]
params_linear =[]
params_kdtree =[]
mydata_loaded =[]

@app.task
def upload_data(**kwargs):
    global upload_nums
    global params_kdtree
    global params_linear
    global features_train
    global flann_linear
    global flann_kdtree
    global mydata_loaded

    json_dump=kwargs['json_dump']
    json_load = json.loads(json_dump)
    mydata_loaded = np.asarray(json_load["mydata"])
    print(len(mydata_loaded))
    print(mydata_loaded[0]['path'])

    print('*** data uploaded ***')

    dim=len(mydata_loaded[0]['features'])
    features_train=np.zeros([len(mydata_loaded),dim])
    for iter in range(len(mydata_loaded)):
        features_train[iter,:] = mydata_loaded[iter]['features']

    # ****************************************
    #    *** you will need to complete this part ***
    # Create a near neighbour data structure here from the feature_train feature vectors!
    # ********************************************
    params = flann_kdtree.build_index(features_train, algorithm="kdtree", target_precision=0.9, log_level = "info")
    params_kdtree.append(params)

    return params

@app.task
def imagesearch_tasks(**kwargs):
    global upload_nums
    global params_kdtree
    global params_linear
    global features_train
    global flann_linear
    global flann_kdtree
    global mydata_loaded

    num_results=5

    json_dump=kwargs['json_dump']
    json_load = json.loads(json_dump)
    query_feature = np.asarray(json_load["query_feature"])

    results=[]

    # ****************************************
    #    *** you will need to complete this part ***
    # Find k nearest neighbours that are closest to your query vector
    # ********************************************

    params = params_kdtree[0]
    start_time = time.time()
    result, dists = flann_kdtree.nn_index(query_feature, num_results, checks=params["checks"]);
    time_taken = time.time() - start_time
    return [result[0].tolist(), time_taken]

@app.task
def bruteForce(**kwargs):

    global upload_nums
    global params_kdtree
    global params_linear
    global features_train
    global flann_linear
    global flann_kdtree
    global mydata_loaded

    json_dump=kwargs['json_dump']
    json_load = json.loads(json_dump)
    mydata_loaded = np.asarray(json_load["mydata"])
    query_feature = np.asarray(json_load["query_feature"])

    print(len(mydata_loaded))
    print(mydata_loaded[0]['path'])

    print('*** data uploaded ***')

    dim=len(mydata_loaded[0]['features'])
    features_train=np.zeros([len(mydata_loaded),dim])

    num_results = 5
    for iter in range(len(mydata_loaded)):
        features_train[iter,:] = mydata_loaded[iter]['features']

    params = flann_linear.build_index(features_train, algorithm= "linear", target_precision=0.9, log_level = "info")

    start_time = time.time()
    result, dists = flann_linear.nn_index(query_feature, num_results, checks=params["checks"]);
    time_taken = time.time() - start_time
    return [result[0].tolist(), time_taken]

# Euclidean Distance Caculator
def dist(a, b, ax=1):
    return np.linalg.norm(a - b, axis=ax)

class NumpyEncoder(json.JSONEncoder):
    def default(self, obj):
        if isinstance(obj, np.ndarray):
            return obj.tolist()
        elif isinstance(obj, np.int64):
            return int(obj)
        return json.JSONEncoder.default(self, obj)