Forked from robertodaza/prueba1/3
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.
| Endpoint Name | Data Format | Nature |
|---|---|---|
| comparison_ids | system/array_1d_text/1 | Input |
| keystroke | tutorial/atvs_keystroke/1 | Input |
| probe_client_id | system/text/1 | Input |
| scores | elie_khoury/string_probe_scores/1 | Output |
| Endpoint Name | Data Format | Nature |
|---|---|---|
| template_client_id | system/text/1 | Input |
| id | system/text/1 | Input |
| features | tutorial/atvs_keystroke/1 | Input |
Parameters allow users to change the configuration of an algorithm when scheduling an experiment
| Name | Description | Type | Default | Range/Choices |
|---|---|---|---|---|
| field | Data field used to generate the feature template | string | given_name | given_name, family_name, email, nationality, id_number, all_five |
| distance | Distance to obtain the matching score | string | Modified Scaled Manhattan | Scaled Manhattan, Modified Scaled Manhattan, Combined Manhattan-Mahalanobis, Mahalanobis + Nearest Neighbor |
Algorithms may use functions and classes
declared in libraries. Here you can see the libraries and
import names used by this library. You don't
need to import the library manually on your code, the platform
will do it for you. Just use the object as it has been imported
with the selected named. For example, if you choose to import a
library using the name lib, then access function
f within your code like lib.f().
| Library | Import as |
|---|---|
| robertodaza/competition--modified_scaled_distance--scaled_distance--mad-/1 | prueba |
xxxxxxxxxximport numpy#en la libreria robertodaza/competition--modified_scaled_distance--scaled_distance--mad-/1 que la hemos nombrado como prueba tenemos todas las funciones que usaremosclass Algorithm: def __init__(self):#primer modulo donde definimos las variables globales self._features1 = [] #definimos 5 listas self._features2 = [] self._features3 = [] self._features4 = [] self._features5 = [] self._templates=None#lo ponemoscon valor none self._field='given_name' #Por defecto el nombre self._distance='Modified Scaled Manhattan'#Por defecto modified scaled manhattan def setup(self, parameters): #Funcion que define los parametros (PODEMOS CAMBIAR EN PARAMETRO EL VALOR DE FIELD Y DISTANCE CON ESTA FUNCION) self._field = parameters.get('field', self._field)#si el parametro field no toma valor se deja el nombre si toma valor se pondra el que sea(correo,DNI,etc) self._distance = parameters.get('distance', self._distance)#igual pero con distance return True def process(self, inputs, outputs):#trabajamos con objetos if self._templates is None:#La primera vez entra aqui self._templates = {}#define un diccionario group = inputs.groupOf('features')#como tenemos un bloque que le entran entradas de dos bloques diferentes y no estan sincronizados tenemos que hacer grupos, con esta funcion abrimos un grupo while group.hasMoreData():#Mientras tenga datos se ejecuta group.next()# pedir dato nuevo del grupo f1 = group['features'].data.holdtime.given_name#abrimos features que es las muestras que nos dan de la base de datos y cogemos de holdtime el nombre f2 = group['features'].data.rplatency.given_name#igual pero con rplatency feature_vector=[]#lista l_f=len(f1)#es la longitud son caracteres for i_k in range(l_f):#hacemos un for para ponerlos en el vector features_vector transformando a float feature_vector.append(float(f1[i_k])) l_f=len(f2)#igual for i_k in range(l_f): feature_vector.append(float(f2[i_k])) self._features1.append(feature_vector)#vamos haciendo una lista que es una matriz con cada prueba de las personas asi que seran 6 filas f1 = group['features'].data.holdtime.family_name#igual con family f2 = group['features'].data.rplatency.family_name feature_vector=[] l_f=len(f1) for i_k in range(l_f): feature_vector.append(float(f1[i_k])) l_f=len(f2) for i_k in range(l_f): feature_vector.append(float(f2[i_k])) self._features2.append(feature_vector) f1 = group['features'].data.holdtime.email#igual con email f2 = group['features'].data.rplatency.email feature_vector=[] l_f=len(f1) for i_k in range(l_f): feature_vector.append(float(f1[i_k])) l_f=len(f2) for i_k in range(l_f): feature_vector.append(float(f2[i_k])) self._features3.append(feature_vector) f1 = group['features'].data.holdtime.nationality#igual con nationality f2 = group['features'].data.rplatency.nationality feature_vector=[] l_f=len(f1) for i_k in range(l_f): feature_vector.append(float(f1[i_k])) l_f=len(f2) for i_k in range(l_f): feature_vector.append(float(f2[i_k])) self._features4.append(feature_vector) f1 = group['features'].data.holdtime.id_number#igual con id_number f2 = group['features'].data.rplatency.id_number feature_vector=[] l_f=len(f1) for i_k in range(l_f): feature_vector.append(float(f1[i_k])) l_f=len(f2) for i_k in range(l_f): feature_vector.append(float(f2[i_k])) self._features5.append(feature_vector) # Calculate mean and std if inputs["id"].isDataUnitDone():#id es el numero por el usuario por el que va, si ha terminado con el usuario entra y realiza lo siguiente features11 = numpy.average(self._features1, axis=0)#calcula la media quedandote solo una fila, media por columnas features12 = numpy.std(self._features1, axis=0)#std igual features13 = prueba.mad(self._features1)#mad igual features21 = numpy.average(self._features2, axis=0) features22 = numpy.std(self._features2, axis=0) features23 = prueba.mad(self._features2) features31 = numpy.average(self._features3, axis=0) features32 = numpy.std(self._features3, axis=0) features33 = prueba.mad(self._features3) features41 = numpy.average(self._features4, axis=0) features42 = numpy.std(self._features4, axis=0) features43 = prueba.mad(self._features4) features51 = numpy.average(self._features5, axis=0) features52 = numpy.std(self._features5, axis=0) features53 = prueba.mad(self._features5) template_id = group['id'].data.text#miramos por el numero por el que va if self._field == 'all_five':# si es con todos self._templates[template_id] = dict( #creamos un diccionario y guardamos en cada posicion del usuario por el que va los valores client_id = group['template_client_id'].data.text,#guardamos el nombre del usuario por el que va ejemplo(Genuine_12) model_average1 = features11, model_std1 = features12, model_mad1 = features13, model_average2= features21, model_std2= features22, model_mad2= features23, model_average3= features31, model_std3= features32, model_mad3= features33, model_average4= features41, model_std4= features42, model_mad4= features43, model_average5= features51, model_std5= features52, model_mad5= features53, ) if self._field == 'given_name':#si es el nombre hacemos lo mismo que antes pero solo guardamos los datos del nombre self._templates[template_id] = dict( client_id = group['template_client_id'].data.text, model_average = features11, model_std = features12, model_mad= features13, ) if self._field == 'family_name':#igual self._templates[template_id] = dict( client_id = group['template_client_id'].data.text, model_average= features21, model_std= features22, model_mad= features23, ) if self._field == 'email':#igual self._templates[template_id] = dict( client_id = inputs['template_client_id'].data.text, model_average= features31, model_std= features32, model_mad= features33, ) if self._field == 'nationality':#igual self._templates[template_id] = dict( client_id = inputs['template_client_id'].data.text, model_average= features41, model_std= features42, model_mad= features43, ) if self._field == 'id_number':#igual self._templates[template_id] = dict( client_id = inputs['template_client_id'].data.text, model_average= features51, model_std= features52, model_mad= features53, ) self._features1 = []#cuando termina ponemos las listas a cero para volver a empezar self._features2 = [] self._features3 = [] self._features4 = [] self._features5 = [] comparison_ids = inputs['comparison_ids'].data.text#cuando ha terminado con el grupo de templates empieza con probe, aqui guardamos el numero por el que va de las de prueba data = inputs['keystroke'].data #son los datos de las muestras de prueba if self._field == 'given_name':#si es el nombre coge el hotdtime y rplatency f1 = data.holdtime.given_name f2 = data.rplatency.given_name if self._field == 'family_name':#igual f1 = data.holdtime.family_name f2 = data.rplatency.family_name if self._field == 'email':#igual f1 = data.holdtime.email f2 = data.rplatency.email if self._field == 'nationality': f1 = data.holdtime.nationality f2 = data.rplatency.nationality if self._field == 'id_number': f1 = data.holdtime.id_number f2 = data.rplatency.id_number if self._field == 'all_five':#Si es todo coge los cinco f11 = data.holdtime.given_name f12 = data.rplatency.given_name f21 = data.holdtime.family_name f22 = data.rplatency.family_name f31 = data.holdtime.email f32 = data.rplatency.email f41 = data.holdtime.nationality f42 = data.rplatency.nationality f51 = data.holdtime.id_number f52 = data.rplatency.id_number if self._field != 'all_five':#si es distinto de los 5 pasw solo lo hace para uno feature_vector=[] #definimos una lista l_f=len(f1)#longitud de la lista for i_k in range(l_f):#vamos poniendo los valores en una fila y transformando a float feature_vector.append(float(f1[i_k])) l_f=len(f2) for i_k in range(l_f):#igual pero con rplatency feature_vector.append(float(f2[i_k])) probe_features=feature_vector#igualamos el vector if self._field == 'all_five':#con los cindo si es all_five feature_vector1=[] l_f=len(f11) for i_k in range(l_f): feature_vector1.append(float(f11[i_k])) l_f=len(f12) for i_k in range(l_f): feature_vector1.append(float(f12[i_k])) feature_vector2=[] l_f=len(f21) for i_k in range(l_f): feature_vector2.append(float(f21[i_k])) l_f=len(f22) for i_k in range(l_f): feature_vector2.append(float(f22[i_k])) feature_vector3=[] l_f=len(f31) for i_k in range(l_f): feature_vector3.append(float(f31[i_k])) l_f=len(f32) for i_k in range(l_f): feature_vector3.append(float(f32[i_k])) feature_vector4=[] l_f=len(f41) for i_k in range(l_f): feature_vector4.append(float(f41[i_k])) l_f=len(f42) for i_k in range(l_f): feature_vector4.append(float(f42[i_k])) feature_vector5=[] l_f=len(f51) for i_k in range(l_f): feature_vector5.append(float(f51[i_k])) l_f=len(f52) for i_k in range(l_f): feature_vector5.append(float(f52[i_k])) probe_features1=feature_vector1#tenemos las cinco muestras de una fila probe_features2=feature_vector2 probe_features3=feature_vector3 probe_features4=feature_vector4 probe_features5=feature_vector5 scores = []#creamos una lista for comparison_id in comparison_ids:#for con el numero de usuario de prueba por el que va template_client_identity = self._templates[comparison_id]['client_id'] #guardamos el nombre del usuario por el que va if self._field != 'all_five':#Hacemos las distancias para solo un pasw if self._distance == 'Modified Scaled Manhattan': score=prueba.modified_scaled_distance(self._templates[comparison_id]['model_average'],self._templates[comparison_id]['model_std'],probe_features); if self._distance == 'Scaled Manhattan': score=prueba.scaled_distance(self._templates[comparison_id]['model_average'],self._templates[comparison_id]['model_mad'],probe_features); if self._field == 'all_five':#para los cinco pasw score=[] if self._distance == 'Modified Scaled Manhattan': score.append(prueba.modified_scaled_distance(self._templates[comparison_id]['model_average1'],self._templates[comparison_id]['model_std1'],probe_features1)) score.append(prueba.modified_scaled_distance(self._templates[comparison_id]['model_average2'],self._templates[comparison_id]['model_std2'],probe_features2)) score.append(prueba.modified_scaled_distance(self._templates[comparison_id]['model_average3'],self._templates[comparison_id]['model_std3'],probe_features3)) score.append(prueba.modified_scaled_distance(self._templates[comparison_id]['model_average4'],self._templates[comparison_id]['model_std4'],probe_features4)) score.append(prueba.modified_scaled_distance(self._templates[comparison_id]['model_average5'],self._templates[comparison_id]['model_std5'],probe_features5)) if self._distance == 'Scaled Manhattan': score.append(prueba.scaled_distance(self._templates[comparison_id]['model_average1'],self._templates[comparison_id]['model_std1'],probe_features1)) score.append(prueba.scaled_distance(self._templates[comparison_id]['model_average2'],self._templates[comparison_id]['model_std2'],probe_features2)) score.append(prueba.scaled_distance(self._templates[comparison_id]['model_average3'],self._templates[comparison_id]['model_std3'],probe_features3)) score.append(prueba.scaled_distance(self._templates[comparison_id]['model_average4'],self._templates[comparison_id]['model_std4'],probe_features4)) score.append(prueba.scaled_distance(self._templates[comparison_id]['model_average5'],self._templates[comparison_id]['model_std5'],probe_features5)) score=numpy.average(score)#media de todos los scores por fila scores.append({#creamos una variable que tiene dos valores dentro score y el usuario genuine 'template_identity': template_client_identity, 'score': score, }) outputs['scores'].write({#escribimos en la salida por el usuario de prueba por el que va, ejemplo impostor_2, y scores 'client_identity': inputs['probe_client_id'].data.text, 'scores': scores }, ) return TrueThe 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 of keystroke data. It enrolls a model from several features by computing the average and standard deviation of the enrollment features.
Note
All features must have the same length.
| Updated | Name | Databases/Protocols | Analyzers | |||
|---|---|---|---|---|---|---|
| robertodaza/robertodaza/example2/2/article_one_block | atvskeystroke/1@A | robertodaza/analyzercompetition/2 | ||||
| robertodaza/robertodaza/example2/2/article_one_block1 | atvskeystroke/1@A | robertodaza/analyzerahora/1 |
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.