1. algorithms
  2. aythamimm
  3. enroll_keystroke
  4. 36

Enrolls a model by averaging all enrollment features

This algorithm is a legacy one. The API has changed since its implementation. New versions and forks will need to be updated.
This algorithm is splittable

Algorithms have at least one input and one output. All algorithm endpoints are organized in groups. Groups are used by the platform to indicate which inputs and outputs are synchronized together. The first group is automatically synchronized with the channel defined by the block in which the algorithm is deployed.

Group: main

Endpoint Name Data Format Nature
keystroke tutorial/atvs_keystroke/1 Input
template_id system/text/1 Input
keystroke_model aythamimm/keystroke_model/6 Output 
xxxxxxxxxx
115
 
1
import numpy
2
​
3
class Algorithm:
4
​
5
    def __init__(self):
6
        self._features1 = []
7
        self._features2 = []
8
        self._features3 = []
9
        self._features4 = []
10
        self._features5 = []        
11
​
12
    def process(self, inputs, outputs):
13
​
14
        # collect all the features for the current template
15
        data = inputs['keystroke'].data
16
      
17
        f1 = data.holdtime.given_name
18
        f2 = data.rplatency.given_name
19
        feature_vector=[]
20
        l_f=len(f1)
21
        for i_k in range(l_f):
22
            feature_vector.append(float(f1[i_k]))
23
      
24
        l_f=len(f2)
25
        for i_k in range(l_f):
26
            feature_vector.append(float(f2[i_k]))            
27
              
28
        self._features1.append(feature_vector)
29
                    
30
        f1 = data.holdtime.family_name
31
        f2 = data.rplatency.family_name
32
        feature_vector=[]
33
        l_f=len(f1)
34
        for i_k in range(l_f):
35
            feature_vector.append(float(f1[i_k]))
36
      
37
        l_f=len(f2)
38
        for i_k in range(l_f):
39
            feature_vector.append(float(f2[i_k]))            
40
              
41
        self._features2.append(feature_vector)
42
              
43
        f1 = data.holdtime.email
44
        f2 = data.rplatency.email
45
        feature_vector=[]
46
        l_f=len(f1)
47
        for i_k in range(l_f):
48
            feature_vector.append(float(f1[i_k]))
49
      
50
        l_f=len(f2)
51
        for i_k in range(l_f):
52
            feature_vector.append(float(f2[i_k]))            
53
              
54
        self._features3.append(feature_vector)
55
      
56
        f1 = data.holdtime.nationality
57
        f2 = data.rplatency.nationality
58
        feature_vector=[]
59
        l_f=len(f1)
60
        for i_k in range(l_f):
61
            feature_vector.append(float(f1[i_k]))
62
      
63
        l_f=len(f2)
64
        for i_k in range(l_f):
65
            feature_vector.append(float(f2[i_k]))            
66
              
67
        self._features4.append(feature_vector)
68
      
69
        f1 = data.holdtime.id_number
70
        f2 = data.rplatency.id_number
71
        feature_vector=[]
72
        l_f=len(f1)
73
        for i_k in range(l_f):
74
            feature_vector.append(float(f1[i_k]))
75
      
76
        l_f=len(f2)
77
        for i_k in range(l_f):
78
            feature_vector.append(float(f2[i_k]))            
79
              
80
        self._features5.append(feature_vector)
81
              
82
        # Calculate mean and std
83
        if inputs["template_id"].isDataUnitDone():
84
            features11 = numpy.average(self._features1, axis=0)
85
            features12 = numpy.std(self._features1, axis=0)
86
            features21 = numpy.average(self._features2, axis=0)
87
            features22 = numpy.std(self._features2, axis=0)    
88
            features31 = numpy.average(self._features3, axis=0)
89
            features32 = numpy.std(self._features3, axis=0)    
90
            features41 = numpy.average(self._features4, axis=0)
91
            features42 = numpy.std(self._features4, axis=0)    
92
            features51 = numpy.average(self._features5, axis=0)
93
            features52 = numpy.std(self._features5, axis=0)    
94
            # outputs data
95
            outputs["keystroke_model"].write({
96
                'given_name_average':       features11,
97
                'given_name_std':           features12,
98
                'family_name_average':      features21,
99
                'family_name_std':          features22,
100
                'email_average':            features31,
101
                'email_std':                features32,
102
                'nationality_average':      features41,
103
                'nationality_std':          features42,
104
                'id_number_average':        features51,
105
                'id_number_std':            features52,
106
            })
107
          
108
            self._features1 = []
109
            self._features2 = []
110
            self._features3 = []
111
            self._features4 = []
112
            self._features5 = []
113
​
114
        return True
115
​

The code for this algorithm in Python
The ruler at 80 columns indicate suggested POSIX line breaks (for readability).
The editor will automatically enlarge to accomodate the entirety of your input
Use keyboard shortcuts for search/replace and faster editing. For example, use Ctrl-F (PC) or Cmd-F (Mac) to search through this box

This algorithm is designed to be used as a simple enrollment strategy. It enrolls a model from several features by computing the average of the enrollment features.

Note

All features must have the same length.

Experiments

Updated Name Databases/Protocols Analyzers
aythamimm/aythamimm/btas15_keystroke_experiments/6/BTAS_2015_Kesytroke_Experiment atvskeystroke/1@A aythamimm/analyzer_keystroke/70
Created with Raphaël 2.1.2[compare]aythamimm/enroll_keystroke/36robertodaza/enroll-keystroke-proof1-commented/1Jul2015Dec28

This table shows the number of times this algorithm has been successfully run using the given environment. Note this does not provide sufficient information to evaluate if the algorithm will run when submitted to different conditions.

Terms of Service | Contact Information | BEAT platform version 2.2.1b0 | © Idiap Research Institute - 2013-2025