Delete titanic_rn_2024_2.py

titanic_rn_2024_2.py

@@ -1,155 +0,0 @@
-# -*- coding: utf-8 -*-
-"""Titanic_RN.ipynb
-
-Automatically generated by Colab.
-
-Original file is located at
-    https://colab.research.google.com/drive/1bBTik7AiMvb-_MLAf5rTR0STa8nfJp-d
-"""
-
-# importar la libreria
-import pandas as pd
-
-"""Crear el dataframe"""
-
-training = pd.read_csv("titanic-train.csv")
-training.head(5)
-
-"""Verificar el cintenido del dataframe training"""
-
-training.info()
-
-"""Reemplazar los datos del genero a entero
-male==> 0           female==> 1
-"""
-
-training['Gender'] = training['Gender'].apply(lambda toLabel:0 if toLabel=='male' else 1)
-
-training.head(5)
-
-"""Completar los valores Nan correspondiente a la edad ( Age ) con el promedio de edades validos"""
-
-training["Age"].fillna(training["Age"].mean(),inplace=True)
-
-training.head(5)
-
-"""Verrificar los datos del dataset"""
-
-training.info()
-
-"""Indentificar los input data y los target data"""
-
-y_target = training["Survived"].values
-print(y_target)
-
-columns = ["Fare","Pclass","Gender","Age","SibSp"]
-x_input = training[list(columns)].values
-print(x_input)
-
-"""Definir la estructura de la red nuronal para el aprendizaje"""
-
-import keras
-from keras import layers
-from keras import ops
-
-model = keras.Sequential()
-model.add(layers.Dense(16,input_dim=5,activation='relu'))
-model.add(layers.Dense(16, activation="relu", name="layer1"))
-model.add(layers.Dense(1, activation='sigmoid',name="layer2"))
-
-"""Configurar los parametros de la red"""
-
-model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])
-
-"""Realizar el proceso de entrenamiento de la red neuronal"""
-
-model.fit(x_input,y_target,epochs=1000)
-
-"""Evaluar la presicion."""
-
-score = model.evaluate(x_input,y_target)
-print("\n %s: %.2f%%" % (model.metrics_names[1],score[1]*100))
-
-XX = np.array(x_input[[2]])
-print(XX)
-respuesta = model.predict(XX)
-print(respuesta.round()[0][0])
-
-y_simulado = []
-for i in range(len(y_target)):
-  y_simulado.append(model.predict(x_input[[i]]).round())
-print(y_simulado)
-
-import numpy as np
-respuesta = model.predict(np.array([[83.475, 1, 1, 35, 1,]]))
-#respuesta = model.predict(np.array([[ 7.8958, 3.0, 0.0, 29.97086721, 0.0]]))
-#respuesta = model.predict(np.array([[82.2667,1,1,23.000000,1]]))
-print(respuesta.round()[0][0])
-
-precision = 0
-for i in range(len(y_simulado)):
-  if y_target[i]==y_simulado[i].round()[0][0]:
-    precision = precision + 1
-precision = precision/float(len(y_simulado))
-print(precision)
-
-y_simulado_lista = []
-for i in range(len(y_simulado)):
-  y_simulado_lista.append(y_simulado[i].round()[0][0])
-print(y_simulado_lista)
-
-import numpy as np
-from sklearn.metrics import confusion_matrix
-Matriz_de_Confusion = confusion_matrix(y_target, y_simulado_lista)
-Matriz_de_Confusion
-"""sensibilidad"""
-#sensibilidad = (Matriz_de_Confusion[0,0])/np.sum(Matriz_de_Confusion[0,0]+ Matriz_de_Confusion[1,0])
-VP = Matriz_de_Confusion[0,0]
-FN = Matriz_de_Confusion[1,0]
-sensibilidad = VP/(VP+FN)
-sensibilidad
-
-"""especificidad"""
-#especificidad = (Matriz_de_Confusion[1,1])/np.sum(Matriz_de_Confusion[1,1]+ Matriz_de_Confusion[0,1])
-VN = Matriz_de_Confusion[1,1]
-FP = Matriz_de_Confusion[0,1]
-especificidad = VN/(VN+FP)
-especificidad
-
-from sklearn.metrics import confusion_matrix
-from matplotlib import pyplot as plt
-from sklearn.metrics import classification_report
-
-conf_mat = confusion_matrix(y_true=y_target, y_pred=y_simulado_lista)
-print('Matriz de Confusión - DATOS ORIGINALES:\n', conf_mat)
-print('Métricas de Matriz de Confusión - DATOS ORIGINALES:\n',classification_report(y_target,y_simulado_lista))
-labels = ['Class 0', 'Class 1']
-fig = plt.figure()
-ax = fig.add_subplot(111)
-cax = ax.matshow(conf_mat, cmap=plt.cm.Blues)
-fig.colorbar(cax)
-ax.set_xticklabels([''] + labels)
-ax.set_yticklabels([''] + labels)
-plt.xlabel('Predicted')
-plt.ylabel('Expected')
-plt.show()
-
-#Grabando la estructura
-model_json = model.to_json()
-with open("mimodel.json","w") as json_file:
-  json_file.write(model_json)
-
-# Grabando los pesos
-model.save_weights("mimodelo.weights.h5")
-print("modelo guardado")
-
-from keras.models import Model,model_from_json
-json_file = open("model.json",'r')
-loaded_model_json = json_file.read()
-json_file.close()
-loaded_model = model_from_json(loaded_model_json)
-loaded_model.load_weights("mimodelo.weights.h5")
-print("modelo cargado en el disco")
-loaded_model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])
-
-print(loaded_model.predict(np.array([[83.475, 1, 1, 35, 1,]])).round()[0][0])