Feat: Added data

BECK/app.py

@@ -28,7 +28,21 @@
 
 comment_test = preprocesamiento.preprocesamiento_con_ortografia(
     comments_array[0])
-
+array_item = []
 # Get Vector Beck
+item_string = 'Pensamiento o deseos suicidas'
+for key in beck_data_preprocessing[item_string].keys():
+  array_item.append(beck_data_preprocessing[item_string][key]["data"])
+
+print(comment_test)
+i = 0
+for item in array_item:
+  coseno = w2v.get_cosine_distance(item, comment_test) 
+  print(f'{item_string} - Item BECK {i} distancia coseno: ${coseno}')
+  euclidian = w2v.get_euclidian_distance(["hoy", "me", "quiero", "morir"], ["no", "morir"])
+  print(f'{item_string} - Item BECK {i} distancia euclidiana: ${euclidian}')
+  i += 1
+
+
 result = w2v.getVectorBeck(comment_test, beck_data_preprocessing)
 print(result)

BECK/coseno_vs_euclidian.py

@@ -0,0 +1,56 @@
+import pandas as pd
+from preprocessing_service import Preprocesamiento
+import json
+import pprint
+from model_word2vec_service import ModelWord2Vec
+df = pd.read_excel("./datasets/DATASET_ENTRENAMIENTO.xlsx",index_col=[1,2]).reset_index()
+comments = list(df["text"])
+classes =  list(df["class"])
+
+
+# df_prueba_chelsea = df_prueba_chelsea.append({'Nombre': 'Vaughn', 'Numero de camiseta':'33'}, ignore_index=True)
+comment_0 = comments[0]
+preprocesamiento = Preprocesamiento()
+
+w2v = ModelWord2Vec()
+df_cve = pd.read_csv('./coseno_vs_euclidian.csv')
+columns = list(df_cve.columns)[2:]
+
+# Lectura beck
+beck_data_preprocessing = {}
+try:
+    if open('./JSON/items_preprocessing.json', 'r'):
+        beck_data_preprocessing = json.loads(
+            open('./JSON/items_preprocessing.json', 'r', encoding='utf-8').read())
+except Exception as e:
+    print(f'Error: {e}')
+
+class_comment = 0
+for comment in comments:
+  try:
+    print(f'Comentario: {class_comment + 1}/{len(comments)}')
+    new_comment = {}
+    contador = 0
+    new_comment["Comentario"] = comment
+    comment_preprocesado = preprocesamiento.preprocesamiento_sin_ortografia(
+        comment)
+    new_comment["Comentario Preprocesado"] = comment_preprocesado
+    w2v.add_corpus(comment_preprocesado)
+    for item in beck_data_preprocessing.keys():
+      for result in beck_data_preprocessing[item].keys():
+        new_comment[columns[contador]] = w2v.get_cosine_distance(comment_preprocesado, beck_data_preprocessing[item][result]["data"])
+        contador += 1
+        new_comment[columns[contador]] = w2v.get_euclidian_distance(comment_preprocesado, beck_data_preprocessing[item][result]["data"])
+        # new_comment[columns[contador]] = 0
+        contador += 1
+
+    # Add to dataframe
+    new_comment["Clase"] =  classes[class_comment]
+    df_cve = df_cve.append(new_comment, ignore_index=True)
+    class_comment += 1
+  except Exception: 
+    print(f'Error en el comentario {class_comment} omitiendo...')
+    class_comment += 1
+    continue
+
+df_cve.to_csv('test.csv', index=False, encoding="utf-8")

BECK/model_word2vec_service.py

@@ -1,3 +1,4 @@
+from pprint import pprint
 from gensim.models import Word2Vec
 import numpy as np
 
@@ -42,22 +43,27 @@ def get_euclidian_distance(self, corpus_a, corpus_b):
     :param corpus_b: El segundo corpus a comparar
     :return: La distancia euclidiana entre dos vectores.
     """
-    if len(corpus_a) !=len(corpus_b) :
-      diferencia= -(len(corpus_a) -len(corpus_b)) 
-      if len(corpus_a) > len(corpus_b):
+    vector_corpus_a = self.get_word_vector(corpus_a)
+    vector_corpus_a = list(np.array(vector_corpus_a).tolist())
+    vector_corpus_b = self.get_word_vector(corpus_b)
+    vector_corpus_b = list(np.array(vector_corpus_b).tolist())
+    #Diferencia entre ambos
+    if len(vector_corpus_a) != len( vector_corpus_b):
+      diferencia = abs(len(vector_corpus_a) - len( vector_corpus_b)) 
+      if len(vector_corpus_a) > len( vector_corpus_b):
             i = 0
             while i < diferencia:
-              corpus_b.append(0)
+              vector_corpus_b.append(self.getVector250())
               i += 1
-            return np.linalg.norm(self.model.wv[corpus_a] - self.model.wv[corpus_b])
-      if len(corpus_a) < len(corpus_b):
+            return np.linalg.norm(np.array(vector_corpus_a) - np.array(vector_corpus_b))
+      if len(vector_corpus_a) < len(vector_corpus_b):
             i = 0
             while i < diferencia:
-              corpus_a.append(0)
+              vector_corpus_a.append(self.getVector250())
               i += 1
-            return np.linalg.norm(self.model.wv[corpus_a] - self.model.wv[corpus_b])
+            return np.linalg.norm(np.array(vector_corpus_a) - np.array(vector_corpus_b))
     else:
-      return np.linalg.norm(self.model.wv[corpus_a] - self.model.wv[corpus_b])
+      return np.linalg.norm(np.array(vector_corpus_a) - np.array(vector_corpus_b))
 
 
   def get_cosine_distance(self, corpus_a, corpus_b):
@@ -90,4 +96,10 @@ def getVectorBeck(self, commentVector, beck):
           itemBeck = beck[item][result]
       array.append(itemBeck['value'])   
     return array
+
+  def getVector250(self):
+    return list(np.zeros(250))
+
+
+