Implementing different ML techniques to improve the prediction results.
Prabhat Turlapati
import numpy as np
import pandas as pd
from sklearn.neighbors import KNeighborsClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.svm import SVC
from sklearn.grid_search import GridSearchCV
from sklearn.metrics import accuracy_score
from sklearn.ensemble import ExtraTreesClassifier
# Use Keras simple neural network for prediction
from keras.models import Sequential
from keras.layers import Dense
# fix random seed for reproducibility
np.random.seed(7)train_raw = pd.read_csv("train.csv")
test_raw = pd.read_csv("test.csv")from IPython.display import Image
Image("picture.png")
# remove un-necessary columns ticket,fare,cabin number,
columns =['PassengerId','Ticket','Fare','Cabin','Name','Embarked']
# train.info()
train = train_raw.drop(columns=columns)
test = test_raw.drop(columns=columns)train.head()
train = pd.get_dummies(data = train, columns=['Sex'])
# train = pd.get_dummies(data = train, columns=['Embarked'])
train = pd.get_dummies(data = train, columns=['Pclass'])
train.drop(columns=['Sex_female'],inplace=True)
test = pd.get_dummies(data = test, columns=['Sex'])
# test = pd.get_dummies(data = test, columns=['Embarked'])
test = pd.get_dummies(data = test, columns=['Pclass'])
test.drop(columns=['Sex_female'],inplace=True)print(train.info())
print(test.info())<class 'pandas.core.frame.DataFrame'>
RangeIndex: 891 entries, 0 to 890
Data columns (total 8 columns):
Survived 891 non-null int64
Age 714 non-null float64
SibSp 891 non-null int64
Parch 891 non-null int64
Sex_male 891 non-null uint8
Pclass_1 891 non-null uint8
Pclass_2 891 non-null uint8
Pclass_3 891 non-null uint8
dtypes: float64(1), int64(3), uint8(4)
memory usage: 31.4 KB
None
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 418 entries, 0 to 417
Data columns (total 7 columns):
Age 332 non-null float64
SibSp 418 non-null int64
Parch 418 non-null int64
Sex_male 418 non-null uint8
Pclass_1 418 non-null uint8
Pclass_2 418 non-null uint8
Pclass_3 418 non-null uint8
dtypes: float64(1), int64(2), uint8(4)
memory usage: 11.5 KB
None
print(train.loc[np.isnan(train['Age'])].head(5)) Survived Age SibSp Parch Sex_male Pclass_1 Pclass_2 Pclass_3
5 0 NaN 0 0 1 0 0 1
17 1 NaN 0 0 1 0 1 0
19 1 NaN 0 0 0 0 0 1
26 0 NaN 0 0 1 0 0 1
28 1 NaN 0 0 0 0 0 1
# Imputing with mean for age
train.Age = train.Age.fillna(value=train.Age.mean())
test.Age = test.Age.fillna(value=test.Age.mean())print(train.head())
print(test.head()) Survived Age SibSp Parch Sex_male Pclass_1 Pclass_2 Pclass_3
0 0 22.0 1 0 1 0 0 1
1 1 38.0 1 0 0 1 0 0
2 1 26.0 0 0 0 0 0 1
3 1 35.0 1 0 0 1 0 0
4 0 35.0 0 0 1 0 0 1
Age SibSp Parch Sex_male Pclass_1 Pclass_2 Pclass_3
0 34.5 0 0 1 0 0 1
1 47.0 1 0 0 0 0 1
2 62.0 0 0 1 0 1 0
3 27.0 0 0 1 0 0 1
4 22.0 1 1 0 0 0 1
y_train = train.Survived
X_train = train.drop(columns=['Survived'])
X_test = test
y_test_raw = pd.read_csv("gender_submission.csv")
y_test = y_test_raw.drop(columns='PassengerId')# Use KNN
knn_clf = KNeighborsClassifier()
gsv_knn = GridSearchCV(knn_clf,{'n_neighbors':[1,3,5]},cv=3,n_jobs=-1)
gsv_knn.fit(X_train,y_train)
y_pred_knn = gsv_knn.predict(X_test)
acc_knn = accuracy_score(y_pred_knn,y_test)
print("knn acc: ",acc_knn)
# Use DT
dt_clf = DecisionTreeClassifier()
gsv_dt = GridSearchCV(estimator=dt_clf,cv=3,n_jobs=-1,param_grid={'max_depth':[10,100,200,250,500,1000]})
gsv_dt.fit(X_train,y_train)
y_pred_dt = gsv_dt.predict(X_test)
acc_dt = accuracy_score(y_pred_dt,y_test)
print("dt acc: ",acc_dt)
# Use RF DT
rf_clf = RandomForestClassifier()
gsv_rf = GridSearchCV(estimator=rf_clf,cv=3,n_jobs=-1,param_grid={'n_estimators':[10,50,100,1000,10000]})
gsv_rf.fit(X_train,y_train)
y_pred_rf = gsv_rf.predict(X_test)
acc_rf = accuracy_score(y_pred_rf,y_test)
print("rf acc: ",acc_rf)
knn acc: 0.7488038277511961
dt acc: 0.7990430622009569
rf acc: 0.777511961722488
# Use SVC
svc_clf = SVC(kernel='linear')
gsv_svc = GridSearchCV(estimator=svc_clf,cv=3,n_jobs=-1,param_grid={'C':np.logspace(-4,0,1, 2, 3)})
gsv_svc.fit(X_train,y_train)
y_pred_svc = gsv_svc.predict(X_test)
acc_svc = accuracy_score(y_pred_svc,y_test)
print("svc acc: ",acc_svc)svc acc: 0.9976076555023924
# Use Extremely Randomized forest Classifier
etc_clf = ExtraTreesClassifier(criterion='gini')
gsv_etc = GridSearchCV(estimator=etc_clf,cv=3,n_jobs=-1,param_grid={'n_estimators':[10,50,100,1000,10000]})
gsv_etc.fit(X_train,y_train)
y_pred_etc = gsv_etc.predict(X_test)
acc_etc = accuracy_score(y_pred_etc,y_test)
print("etc acc: ",acc_etc)etc acc: 0.7607655502392344
model = Sequential()
model.add(Dense(12, input_dim=len(X_train.columns), activation='relu'))
model.add(Dense(8, activation='relu'))
model.add(Dense(1, activation='sigmoid'))model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])model.fit(X_train, y_train, nb_epoch=300, batch_size=30)F:\Anaconda3\lib\site-packages\ipykernel_launcher.py:1: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.
"""Entry point for launching an IPython kernel.
Epoch 1/300
891/891 [==============================] - 0s 368us/step - loss: 1.7415 - acc: 0.6162
Epoch 2/300
891/891 [==============================] - 0s 35us/step - loss: 0.9975 - acc: 0.6173
Epoch 3/300
891/891 [==============================] - 0s 35us/step - loss: 0.7588 - acc: 0.6229
Epoch 4/300
891/891 [==============================] - 0s 35us/step - loss: 0.6652 - acc: 0.6319
Epoch 5/300
891/891 [==============================] - 0s 35us/step - loss: 0.6407 - acc: 0.6364
Epoch 6/300
891/891 [==============================] - 0s 35us/step - loss: 0.6288 - acc: 0.6397
Epoch 7/300
891/891 [==============================] - 0s 35us/step - loss: 0.6189 - acc: 0.6453
Epoch 8/300
891/891 [==============================] - 0s 35us/step - loss: 0.6096 - acc: 0.6611
Epoch 9/300
891/891 [==============================] - 0s 35us/step - loss: 0.5990 - acc: 0.6835
Epoch 10/300
891/891 [==============================] - 0s 35us/step - loss: 0.5899 - acc: 0.7026
Epoch 11/300
891/891 [==============================] - 0s 35us/step - loss: 0.5839 - acc: 0.6947
Epoch 12/300
891/891 [==============================] - 0s 35us/step - loss: 0.5758 - acc: 0.7250
Epoch 13/300
891/891 [==============================] - 0s 35us/step - loss: 0.5705 - acc: 0.7228
Epoch 14/300
891/891 [==============================] - 0s 35us/step - loss: 0.5662 - acc: 0.7452
Epoch 15/300
891/891 [==============================] - 0s 35us/step - loss: 0.5641 - acc: 0.7329
Epoch 16/300
891/891 [==============================] - 0s 18us/step - loss: 0.5568 - acc: 0.7755
Epoch 17/300
891/891 [==============================] - 0s 35us/step - loss: 0.5533 - acc: 0.7699
Epoch 18/300
891/891 [==============================] - 0s 53us/step - loss: 0.5519 - acc: 0.7587
Epoch 19/300
891/891 [==============================] - ETA: 0s - loss: 0.4718 - acc: 0.800 - 0s 18us/step - loss: 0.5473 - acc: 0.7576
Epoch 20/300
891/891 [==============================] - 0s 18us/step - loss: 0.5436 - acc: 0.7699
Epoch 21/300
891/891 [==============================] - 0s 35us/step - loss: 0.5385 - acc: 0.7856
Epoch 22/300
891/891 [==============================] - 0s 35us/step - loss: 0.5354 - acc: 0.7789
Epoch 23/300
891/891 [==============================] - 0s 53us/step - loss: 0.5325 - acc: 0.7811
Epoch 24/300
891/891 [==============================] - 0s 35us/step - loss: 0.5298 - acc: 0.7800
Epoch 25/300
891/891 [==============================] - 0s 35us/step - loss: 0.5279 - acc: 0.7890
Epoch 26/300
891/891 [==============================] - 0s 18us/step - loss: 0.5259 - acc: 0.7834
Epoch 27/300
891/891 [==============================] - 0s 35us/step - loss: 0.5229 - acc: 0.7868
Epoch 28/300
891/891 [==============================] - 0s 53us/step - loss: 0.5200 - acc: 0.7879
Epoch 29/300
891/891 [==============================] - 0s 18us/step - loss: 0.5147 - acc: 0.7823
Epoch 30/300
891/891 [==============================] - 0s 35us/step - loss: 0.5177 - acc: 0.7969
Epoch 31/300
891/891 [==============================] - 0s 35us/step - loss: 0.5105 - acc: 0.7969
Epoch 32/300
891/891 [==============================] - 0s 35us/step - loss: 0.5076 - acc: 0.7957
Epoch 33/300
891/891 [==============================] - 0s 35us/step - loss: 0.5052 - acc: 0.7924
Epoch 34/300
891/891 [==============================] - 0s 35us/step - loss: 0.5058 - acc: 0.8070
Epoch 35/300
891/891 [==============================] - 0s 35us/step - loss: 0.5061 - acc: 0.7991
Epoch 36/300
891/891 [==============================] - 0s 35us/step - loss: 0.4999 - acc: 0.8013
Epoch 37/300
891/891 [==============================] - 0s 35us/step - loss: 0.4984 - acc: 0.8025
Epoch 38/300
891/891 [==============================] - 0s 35us/step - loss: 0.4961 - acc: 0.8047
Epoch 39/300
891/891 [==============================] - 0s 35us/step - loss: 0.4957 - acc: 0.8070
Epoch 40/300
891/891 [==============================] - 0s 35us/step - loss: 0.4933 - acc: 0.7991
Epoch 41/300
891/891 [==============================] - 0s 18us/step - loss: 0.4892 - acc: 0.8070
Epoch 42/300
891/891 [==============================] - 0s 18us/step - loss: 0.4898 - acc: 0.8036
Epoch 43/300
891/891 [==============================] - 0s 18us/step - loss: 0.4894 - acc: 0.8036
Epoch 44/300
891/891 [==============================] - 0s 35us/step - loss: 0.4864 - acc: 0.8092
Epoch 45/300
891/891 [==============================] - 0s 35us/step - loss: 0.4838 - acc: 0.8081
Epoch 46/300
891/891 [==============================] - 0s 35us/step - loss: 0.4820 - acc: 0.8204
Epoch 47/300
891/891 [==============================] - 0s 35us/step - loss: 0.4837 - acc: 0.8092
Epoch 48/300
891/891 [==============================] - 0s 35us/step - loss: 0.4796 - acc: 0.8070
Epoch 49/300
891/891 [==============================] - 0s 35us/step - loss: 0.4840 - acc: 0.8148
Epoch 50/300
891/891 [==============================] - 0s 35us/step - loss: 0.4774 - acc: 0.8137
Epoch 51/300
891/891 [==============================] - 0s 35us/step - loss: 0.4756 - acc: 0.8103
Epoch 52/300
891/891 [==============================] - 0s 18us/step - loss: 0.4761 - acc: 0.8137
Epoch 53/300
891/891 [==============================] - 0s 35us/step - loss: 0.4729 - acc: 0.8204
Epoch 54/300
891/891 [==============================] - 0s 35us/step - loss: 0.4737 - acc: 0.8182
Epoch 55/300
891/891 [==============================] - 0s 35us/step - loss: 0.4715 - acc: 0.8193
Epoch 56/300
891/891 [==============================] - 0s 35us/step - loss: 0.4730 - acc: 0.8182
Epoch 57/300
891/891 [==============================] - 0s 35us/step - loss: 0.4734 - acc: 0.8126
Epoch 58/300
891/891 [==============================] - 0s 35us/step - loss: 0.4732 - acc: 0.8126
Epoch 59/300
891/891 [==============================] - 0s 18us/step - loss: 0.4693 - acc: 0.8171
Epoch 60/300
891/891 [==============================] - 0s 35us/step - loss: 0.4691 - acc: 0.8126
Epoch 61/300
891/891 [==============================] - 0s 35us/step - loss: 0.4672 - acc: 0.8148
Epoch 62/300
891/891 [==============================] - 0s 35us/step - loss: 0.4659 - acc: 0.8148
Epoch 63/300
891/891 [==============================] - 0s 35us/step - loss: 0.4672 - acc: 0.8081
Epoch 64/300
891/891 [==============================] - 0s 35us/step - loss: 0.4700 - acc: 0.8081
Epoch 65/300
891/891 [==============================] - 0s 35us/step - loss: 0.4658 - acc: 0.8148
Epoch 66/300
891/891 [==============================] - 0s 35us/step - loss: 0.4628 - acc: 0.8159
Epoch 67/300
891/891 [==============================] - 0s 18us/step - loss: 0.4616 - acc: 0.8159
Epoch 68/300
891/891 [==============================] - 0s 18us/step - loss: 0.4617 - acc: 0.8204
Epoch 69/300
891/891 [==============================] - 0s 35us/step - loss: 0.4600 - acc: 0.8227
Epoch 70/300
891/891 [==============================] - 0s 35us/step - loss: 0.4609 - acc: 0.8159
Epoch 71/300
891/891 [==============================] - 0s 35us/step - loss: 0.4596 - acc: 0.8215
Epoch 72/300
891/891 [==============================] - 0s 35us/step - loss: 0.4572 - acc: 0.8238
Epoch 73/300
891/891 [==============================] - 0s 35us/step - loss: 0.4603 - acc: 0.8171
Epoch 74/300
891/891 [==============================] - 0s 35us/step - loss: 0.4572 - acc: 0.8182
Epoch 75/300
891/891 [==============================] - 0s 35us/step - loss: 0.4571 - acc: 0.8171
Epoch 76/300
891/891 [==============================] - 0s 18us/step - loss: 0.4565 - acc: 0.8227
Epoch 77/300
891/891 [==============================] - 0s 18us/step - loss: 0.4552 - acc: 0.8182
Epoch 78/300
891/891 [==============================] - 0s 35us/step - loss: 0.4551 - acc: 0.8193
Epoch 79/300
891/891 [==============================] - 0s 35us/step - loss: 0.4583 - acc: 0.8215
Epoch 80/300
891/891 [==============================] - 0s 35us/step - loss: 0.4577 - acc: 0.8204
Epoch 81/300
891/891 [==============================] - 0s 35us/step - loss: 0.4555 - acc: 0.8137
Epoch 82/300
891/891 [==============================] - 0s 35us/step - loss: 0.4552 - acc: 0.8114
Epoch 83/300
891/891 [==============================] - 0s 35us/step - loss: 0.4530 - acc: 0.8193
Epoch 84/300
891/891 [==============================] - 0s 35us/step - loss: 0.4519 - acc: 0.8215
Epoch 85/300
891/891 [==============================] - 0s 35us/step - loss: 0.4545 - acc: 0.8193
Epoch 86/300
891/891 [==============================] - 0s 18us/step - loss: 0.4536 - acc: 0.8193
Epoch 87/300
891/891 [==============================] - 0s 35us/step - loss: 0.4521 - acc: 0.8193
Epoch 88/300
891/891 [==============================] - 0s 35us/step - loss: 0.4517 - acc: 0.8204
Epoch 89/300
891/891 [==============================] - 0s 35us/step - loss: 0.4477 - acc: 0.8193
Epoch 90/300
891/891 [==============================] - 0s 35us/step - loss: 0.4520 - acc: 0.8193
Epoch 91/300
891/891 [==============================] - 0s 35us/step - loss: 0.4483 - acc: 0.8193
Epoch 92/300
891/891 [==============================] - 0s 35us/step - loss: 0.4484 - acc: 0.8182
Epoch 93/300
891/891 [==============================] - 0s 35us/step - loss: 0.4502 - acc: 0.8227
Epoch 94/300
891/891 [==============================] - 0s 18us/step - loss: 0.4557 - acc: 0.8148
Epoch 95/300
891/891 [==============================] - 0s 35us/step - loss: 0.4523 - acc: 0.8171
Epoch 96/300
891/891 [==============================] - 0s 35us/step - loss: 0.4478 - acc: 0.8238
Epoch 97/300
891/891 [==============================] - 0s 35us/step - loss: 0.4454 - acc: 0.8159
Epoch 98/300
891/891 [==============================] - 0s 35us/step - loss: 0.4449 - acc: 0.8204
Epoch 99/300
891/891 [==============================] - 0s 35us/step - loss: 0.4462 - acc: 0.8215
Epoch 100/300
891/891 [==============================] - 0s 35us/step - loss: 0.4465 - acc: 0.8215
Epoch 101/300
891/891 [==============================] - 0s 35us/step - loss: 0.4455 - acc: 0.8171
Epoch 102/300
891/891 [==============================] - 0s 35us/step - loss: 0.4462 - acc: 0.8171
Epoch 103/300
891/891 [==============================] - 0s 18us/step - loss: 0.4426 - acc: 0.8204
Epoch 104/300
891/891 [==============================] - 0s 35us/step - loss: 0.4439 - acc: 0.8260
Epoch 105/300
891/891 [==============================] - 0s 35us/step - loss: 0.4432 - acc: 0.8193
Epoch 106/300
891/891 [==============================] - 0s 35us/step - loss: 0.4502 - acc: 0.8215
Epoch 107/300
891/891 [==============================] - 0s 35us/step - loss: 0.4444 - acc: 0.8171
Epoch 108/300
891/891 [==============================] - 0s 53us/step - loss: 0.4441 - acc: 0.8215
Epoch 109/300
891/891 [==============================] - 0s 35us/step - loss: 0.4435 - acc: 0.8159
Epoch 110/300
891/891 [==============================] - 0s 35us/step - loss: 0.4486 - acc: 0.8148
Epoch 111/300
891/891 [==============================] - 0s 18us/step - loss: 0.4457 - acc: 0.8171
Epoch 112/300
891/891 [==============================] - 0s 35us/step - loss: 0.4425 - acc: 0.8227
Epoch 113/300
891/891 [==============================] - 0s 35us/step - loss: 0.4439 - acc: 0.8193
Epoch 114/300
891/891 [==============================] - 0s 35us/step - loss: 0.4413 - acc: 0.8204
Epoch 115/300
891/891 [==============================] - 0s 35us/step - loss: 0.4410 - acc: 0.8171
Epoch 116/300
891/891 [==============================] - 0s 35us/step - loss: 0.4404 - acc: 0.8182
Epoch 117/300
891/891 [==============================] - 0s 35us/step - loss: 0.4415 - acc: 0.8159
Epoch 118/300
891/891 [==============================] - 0s 18us/step - loss: 0.4431 - acc: 0.8148
Epoch 119/300
891/891 [==============================] - 0s 18us/step - loss: 0.4395 - acc: 0.8260
Epoch 120/300
891/891 [==============================] - 0s 35us/step - loss: 0.4399 - acc: 0.8227
Epoch 121/300
891/891 [==============================] - 0s 35us/step - loss: 0.4407 - acc: 0.8204
Epoch 122/300
891/891 [==============================] - 0s 35us/step - loss: 0.4389 - acc: 0.8204
Epoch 123/300
891/891 [==============================] - 0s 35us/step - loss: 0.4372 - acc: 0.8204
Epoch 124/300
891/891 [==============================] - 0s 18us/step - loss: 0.4395 - acc: 0.8193
Epoch 125/300
891/891 [==============================] - 0s 18us/step - loss: 0.4394 - acc: 0.8204
Epoch 126/300
891/891 [==============================] - 0s 18us/step - loss: 0.4373 - acc: 0.8227
Epoch 127/300
891/891 [==============================] - 0s 35us/step - loss: 0.4379 - acc: 0.8215
Epoch 128/300
891/891 [==============================] - 0s 35us/step - loss: 0.4418 - acc: 0.8182
Epoch 129/300
891/891 [==============================] - 0s 35us/step - loss: 0.4381 - acc: 0.8204
Epoch 130/300
891/891 [==============================] - 0s 35us/step - loss: 0.4353 - acc: 0.8171
Epoch 131/300
891/891 [==============================] - 0s 35us/step - loss: 0.4371 - acc: 0.8193
Epoch 132/300
891/891 [==============================] - 0s 18us/step - loss: 0.4412 - acc: 0.8148
Epoch 133/300
891/891 [==============================] - 0s 35us/step - loss: 0.4376 - acc: 0.8159
Epoch 134/300
891/891 [==============================] - 0s 35us/step - loss: 0.4357 - acc: 0.8238
Epoch 135/300
891/891 [==============================] - 0s 35us/step - loss: 0.4351 - acc: 0.8227
Epoch 136/300
891/891 [==============================] - 0s 35us/step - loss: 0.4359 - acc: 0.8193
Epoch 137/300
891/891 [==============================] - 0s 35us/step - loss: 0.4355 - acc: 0.8204
Epoch 138/300
891/891 [==============================] - 0s 35us/step - loss: 0.4349 - acc: 0.8171
Epoch 139/300
891/891 [==============================] - 0s 35us/step - loss: 0.4364 - acc: 0.8204
Epoch 140/300
891/891 [==============================] - 0s 18us/step - loss: 0.4367 - acc: 0.8193
Epoch 141/300
891/891 [==============================] - 0s 35us/step - loss: 0.4340 - acc: 0.8204
Epoch 142/300
891/891 [==============================] - 0s 35us/step - loss: 0.4334 - acc: 0.8249
Epoch 143/300
891/891 [==============================] - 0s 35us/step - loss: 0.4339 - acc: 0.8171
Epoch 144/300
891/891 [==============================] - 0s 35us/step - loss: 0.4349 - acc: 0.8204
Epoch 145/300
891/891 [==============================] - 0s 35us/step - loss: 0.4326 - acc: 0.8238
Epoch 146/300
891/891 [==============================] - 0s 35us/step - loss: 0.4322 - acc: 0.8204
Epoch 147/300
891/891 [==============================] - 0s 18us/step - loss: 0.4314 - acc: 0.8193
Epoch 148/300
891/891 [==============================] - 0s 35us/step - loss: 0.4342 - acc: 0.8182
Epoch 149/300
891/891 [==============================] - 0s 35us/step - loss: 0.4311 - acc: 0.8171
Epoch 150/300
891/891 [==============================] - 0s 35us/step - loss: 0.4363 - acc: 0.8227
Epoch 151/300
891/891 [==============================] - 0s 35us/step - loss: 0.4348 - acc: 0.8159
Epoch 152/300
891/891 [==============================] - 0s 35us/step - loss: 0.4334 - acc: 0.8182
Epoch 153/300
891/891 [==============================] - 0s 35us/step - loss: 0.4284 - acc: 0.8238
Epoch 154/300
891/891 [==============================] - 0s 35us/step - loss: 0.4322 - acc: 0.8193
Epoch 155/300
891/891 [==============================] - 0s 35us/step - loss: 0.4307 - acc: 0.8215
Epoch 156/300
891/891 [==============================] - 0s 35us/step - loss: 0.4311 - acc: 0.8227
Epoch 157/300
891/891 [==============================] - 0s 18us/step - loss: 0.4306 - acc: 0.8193
Epoch 158/300
891/891 [==============================] - 0s 35us/step - loss: 0.4319 - acc: 0.8215
Epoch 159/300
891/891 [==============================] - 0s 35us/step - loss: 0.4272 - acc: 0.8249
Epoch 160/300
891/891 [==============================] - 0s 35us/step - loss: 0.4288 - acc: 0.8238
Epoch 161/300
891/891 [==============================] - 0s 35us/step - loss: 0.4277 - acc: 0.8204
Epoch 162/300
891/891 [==============================] - 0s 35us/step - loss: 0.4277 - acc: 0.8260
Epoch 163/300
891/891 [==============================] - 0s 35us/step - loss: 0.4309 - acc: 0.8159
Epoch 164/300
891/891 [==============================] - 0s 35us/step - loss: 0.4294 - acc: 0.8238
Epoch 165/300
891/891 [==============================] - 0s 35us/step - loss: 0.4290 - acc: 0.8204
Epoch 166/300
891/891 [==============================] - 0s 18us/step - loss: 0.4262 - acc: 0.8215
Epoch 167/300
891/891 [==============================] - 0s 35us/step - loss: 0.4261 - acc: 0.8249
Epoch 168/300
891/891 [==============================] - 0s 35us/step - loss: 0.4267 - acc: 0.8238
Epoch 169/300
891/891 [==============================] - 0s 35us/step - loss: 0.4264 - acc: 0.8227
Epoch 170/300
891/891 [==============================] - 0s 35us/step - loss: 0.4278 - acc: 0.8238
Epoch 171/300
891/891 [==============================] - 0s 35us/step - loss: 0.4260 - acc: 0.8238
Epoch 172/300
891/891 [==============================] - 0s 35us/step - loss: 0.4273 - acc: 0.8238
Epoch 173/300
891/891 [==============================] - 0s 35us/step - loss: 0.4257 - acc: 0.8249
Epoch 174/300
891/891 [==============================] - 0s 18us/step - loss: 0.4240 - acc: 0.8238
Epoch 175/300
891/891 [==============================] - 0s 35us/step - loss: 0.4247 - acc: 0.8283
Epoch 176/300
891/891 [==============================] - 0s 35us/step - loss: 0.4245 - acc: 0.8272
Epoch 177/300
891/891 [==============================] - 0s 18us/step - loss: 0.4290 - acc: 0.8272
Epoch 178/300
891/891 [==============================] - 0s 35us/step - loss: 0.4241 - acc: 0.8283
Epoch 179/300
891/891 [==============================] - 0s 35us/step - loss: 0.4284 - acc: 0.8249
Epoch 180/300
891/891 [==============================] - 0s 35us/step - loss: 0.4245 - acc: 0.8260
Epoch 181/300
891/891 [==============================] - 0s 35us/step - loss: 0.4271 - acc: 0.8272
Epoch 182/300
891/891 [==============================] - 0s 35us/step - loss: 0.4275 - acc: 0.8249
Epoch 183/300
891/891 [==============================] - 0s 35us/step - loss: 0.4242 - acc: 0.8283
Epoch 184/300
891/891 [==============================] - 0s 18us/step - loss: 0.4247 - acc: 0.8283
Epoch 185/300
891/891 [==============================] - 0s 35us/step - loss: 0.4227 - acc: 0.8272
Epoch 186/300
891/891 [==============================] - 0s 35us/step - loss: 0.4220 - acc: 0.8283
Epoch 187/300
891/891 [==============================] - 0s 35us/step - loss: 0.4257 - acc: 0.8260
Epoch 188/300
891/891 [==============================] - 0s 35us/step - loss: 0.4236 - acc: 0.8272
Epoch 189/300
891/891 [==============================] - 0s 35us/step - loss: 0.4265 - acc: 0.8294
Epoch 190/300
891/891 [==============================] - 0s 35us/step - loss: 0.4253 - acc: 0.8305
Epoch 191/300
891/891 [==============================] - 0s 35us/step - loss: 0.4232 - acc: 0.8249
Epoch 192/300
891/891 [==============================] - 0s 18us/step - loss: 0.4210 - acc: 0.8283
Epoch 193/300
891/891 [==============================] - 0s 35us/step - loss: 0.4220 - acc: 0.8260
Epoch 194/300
891/891 [==============================] - 0s 35us/step - loss: 0.4240 - acc: 0.8238
Epoch 195/300
891/891 [==============================] - 0s 35us/step - loss: 0.4204 - acc: 0.8260
Epoch 196/300
891/891 [==============================] - 0s 35us/step - loss: 0.4210 - acc: 0.8260
Epoch 197/300
891/891 [==============================] - 0s 18us/step - loss: 0.4201 - acc: 0.8305
Epoch 198/300
891/891 [==============================] - 0s 18us/step - loss: 0.4212 - acc: 0.8305
Epoch 199/300
891/891 [==============================] - 0s 35us/step - loss: 0.4210 - acc: 0.8272
Epoch 200/300
891/891 [==============================] - 0s 35us/step - loss: 0.4211 - acc: 0.8283
Epoch 201/300
891/891 [==============================] - 0s 35us/step - loss: 0.4201 - acc: 0.8260
Epoch 202/300
891/891 [==============================] - 0s 35us/step - loss: 0.4185 - acc: 0.8305
Epoch 203/300
891/891 [==============================] - 0s 35us/step - loss: 0.4209 - acc: 0.8283
Epoch 204/300
891/891 [==============================] - 0s 35us/step - loss: 0.4224 - acc: 0.8305
Epoch 205/300
891/891 [==============================] - 0s 35us/step - loss: 0.4206 - acc: 0.8260
Epoch 206/300
891/891 [==============================] - 0s 35us/step - loss: 0.4230 - acc: 0.8204
Epoch 207/300
891/891 [==============================] - 0s 35us/step - loss: 0.4190 - acc: 0.8283
Epoch 208/300
891/891 [==============================] - 0s 35us/step - loss: 0.4248 - acc: 0.8294
Epoch 209/300
891/891 [==============================] - 0s 18us/step - loss: 0.4209 - acc: 0.8283
Epoch 210/300
891/891 [==============================] - 0s 35us/step - loss: 0.4213 - acc: 0.8283
Epoch 211/300
891/891 [==============================] - 0s 35us/step - loss: 0.4203 - acc: 0.8283
Epoch 212/300
891/891 [==============================] - 0s 35us/step - loss: 0.4193 - acc: 0.8294
Epoch 213/300
891/891 [==============================] - 0s 35us/step - loss: 0.4190 - acc: 0.8283
Epoch 214/300
891/891 [==============================] - 0s 35us/step - loss: 0.4212 - acc: 0.8294
Epoch 215/300
891/891 [==============================] - 0s 35us/step - loss: 0.4217 - acc: 0.8272
Epoch 216/300
891/891 [==============================] - ETA: 0s - loss: 0.4674 - acc: 0.800 - 0s 35us/step - loss: 0.4182 - acc: 0.8305
Epoch 217/300
891/891 [==============================] - 0s 35us/step - loss: 0.4182 - acc: 0.8272
Epoch 218/300
891/891 [==============================] - 0s 35us/step - loss: 0.4203 - acc: 0.8249
Epoch 219/300
891/891 [==============================] - 0s 35us/step - loss: 0.4210 - acc: 0.8272
Epoch 220/300
891/891 [==============================] - 0s 35us/step - loss: 0.4180 - acc: 0.8283
Epoch 221/300
891/891 [==============================] - 0s 35us/step - loss: 0.4179 - acc: 0.8283
Epoch 222/300
891/891 [==============================] - 0s 35us/step - loss: 0.4160 - acc: 0.8283
Epoch 223/300
891/891 [==============================] - 0s 18us/step - loss: 0.4184 - acc: 0.8283
Epoch 224/300
891/891 [==============================] - 0s 18us/step - loss: 0.4182 - acc: 0.8294
Epoch 225/300
891/891 [==============================] - 0s 35us/step - loss: 0.4184 - acc: 0.8260
Epoch 226/300
891/891 [==============================] - 0s 35us/step - loss: 0.4197 - acc: 0.8283
Epoch 227/300
891/891 [==============================] - 0s 35us/step - loss: 0.4188 - acc: 0.8283
Epoch 228/300
891/891 [==============================] - 0s 35us/step - loss: 0.4173 - acc: 0.8272
Epoch 229/300
891/891 [==============================] - 0s 35us/step - loss: 0.4168 - acc: 0.8294
Epoch 230/300
891/891 [==============================] - 0s 18us/step - loss: 0.4185 - acc: 0.8294
Epoch 231/300
891/891 [==============================] - 0s 35us/step - loss: 0.4174 - acc: 0.8283
Epoch 232/300
891/891 [==============================] - 0s 18us/step - loss: 0.4176 - acc: 0.8272
Epoch 233/300
891/891 [==============================] - 0s 35us/step - loss: 0.4171 - acc: 0.8294
Epoch 234/300
891/891 [==============================] - 0s 35us/step - loss: 0.4165 - acc: 0.8294
Epoch 235/300
891/891 [==============================] - 0s 35us/step - loss: 0.4165 - acc: 0.8305
Epoch 236/300
891/891 [==============================] - 0s 35us/step - loss: 0.4168 - acc: 0.8260
Epoch 237/300
891/891 [==============================] - 0s 35us/step - loss: 0.4160 - acc: 0.8283
Epoch 238/300
891/891 [==============================] - 0s 35us/step - loss: 0.4173 - acc: 0.8305
Epoch 239/300
891/891 [==============================] - 0s 35us/step - loss: 0.4166 - acc: 0.8260
Epoch 240/300
891/891 [==============================] - 0s 18us/step - loss: 0.4177 - acc: 0.8272
Epoch 241/300
891/891 [==============================] - 0s 35us/step - loss: 0.4175 - acc: 0.8305
Epoch 242/300
891/891 [==============================] - 0s 35us/step - loss: 0.4175 - acc: 0.8294
Epoch 243/300
891/891 [==============================] - 0s 35us/step - loss: 0.4203 - acc: 0.8283
Epoch 244/300
891/891 [==============================] - 0s 53us/step - loss: 0.4218 - acc: 0.8316
Epoch 245/300
891/891 [==============================] - 0s 53us/step - loss: 0.4258 - acc: 0.8249
Epoch 246/300
891/891 [==============================] - 0s 35us/step - loss: 0.4163 - acc: 0.8294
Epoch 247/300
891/891 [==============================] - 0s 36us/step - loss: 0.4195 - acc: 0.8260
Epoch 248/300
891/891 [==============================] - 0s 35us/step - loss: 0.4233 - acc: 0.8227
Epoch 249/300
891/891 [==============================] - 0s 35us/step - loss: 0.4199 - acc: 0.8283
Epoch 250/300
891/891 [==============================] - 0s 35us/step - loss: 0.4190 - acc: 0.8272
Epoch 251/300
891/891 [==============================] - 0s 35us/step - loss: 0.4177 - acc: 0.8283
Epoch 252/300
891/891 [==============================] - 0s 18us/step - loss: 0.4213 - acc: 0.8339
Epoch 253/300
891/891 [==============================] - 0s 18us/step - loss: 0.4220 - acc: 0.8294
Epoch 254/300
891/891 [==============================] - 0s 35us/step - loss: 0.4182 - acc: 0.8305
Epoch 255/300
891/891 [==============================] - 0s 35us/step - loss: 0.4150 - acc: 0.8294
Epoch 256/300
891/891 [==============================] - 0s 35us/step - loss: 0.4167 - acc: 0.8260
Epoch 257/300
891/891 [==============================] - 0s 35us/step - loss: 0.4161 - acc: 0.8283
Epoch 258/300
891/891 [==============================] - 0s 35us/step - loss: 0.4162 - acc: 0.8283
Epoch 259/300
891/891 [==============================] - 0s 18us/step - loss: 0.4144 - acc: 0.8283
Epoch 260/300
891/891 [==============================] - 0s 35us/step - loss: 0.4167 - acc: 0.8305
Epoch 261/300
891/891 [==============================] - 0s 35us/step - loss: 0.4154 - acc: 0.8294
Epoch 262/300
891/891 [==============================] - 0s 35us/step - loss: 0.4186 - acc: 0.8260
Epoch 263/300
891/891 [==============================] - 0s 35us/step - loss: 0.4179 - acc: 0.8260
Epoch 264/300
891/891 [==============================] - 0s 35us/step - loss: 0.4168 - acc: 0.8272
Epoch 265/300
891/891 [==============================] - 0s 18us/step - loss: 0.4141 - acc: 0.8294
Epoch 266/300
891/891 [==============================] - 0s 35us/step - loss: 0.4243 - acc: 0.8283
Epoch 267/300
891/891 [==============================] - 0s 35us/step - loss: 0.4166 - acc: 0.8260
Epoch 268/300
891/891 [==============================] - 0s 35us/step - loss: 0.4164 - acc: 0.8260
Epoch 269/300
891/891 [==============================] - 0s 35us/step - loss: 0.4164 - acc: 0.8294
Epoch 270/300
891/891 [==============================] - 0s 35us/step - loss: 0.4167 - acc: 0.8305
Epoch 271/300
891/891 [==============================] - 0s 35us/step - loss: 0.4149 - acc: 0.8283
Epoch 272/300
891/891 [==============================] - 0s 35us/step - loss: 0.4154 - acc: 0.8272
Epoch 273/300
891/891 [==============================] - 0s 35us/step - loss: 0.4187 - acc: 0.8260
Epoch 274/300
891/891 [==============================] - 0s 35us/step - loss: 0.4159 - acc: 0.8260
Epoch 275/300
891/891 [==============================] - 0s 35us/step - loss: 0.4139 - acc: 0.8294
Epoch 276/300
891/891 [==============================] - 0s 18us/step - loss: 0.4152 - acc: 0.8249
Epoch 277/300
891/891 [==============================] - 0s 18us/step - loss: 0.4159 - acc: 0.8260
Epoch 278/300
891/891 [==============================] - 0s 35us/step - loss: 0.4188 - acc: 0.8238
Epoch 279/300
891/891 [==============================] - 0s 35us/step - loss: 0.4202 - acc: 0.8272
Epoch 280/300
891/891 [==============================] - 0s 35us/step - loss: 0.4174 - acc: 0.8305
Epoch 281/300
891/891 [==============================] - 0s 35us/step - loss: 0.4154 - acc: 0.8238
Epoch 282/300
891/891 [==============================] - 0s 35us/step - loss: 0.4209 - acc: 0.8249
Epoch 283/300
891/891 [==============================] - 0s 35us/step - loss: 0.4181 - acc: 0.8294
Epoch 284/300
891/891 [==============================] - 0s 35us/step - loss: 0.4153 - acc: 0.8272
Epoch 285/300
891/891 [==============================] - 0s 18us/step - loss: 0.4136 - acc: 0.8305
Epoch 286/300
891/891 [==============================] - 0s 35us/step - loss: 0.4175 - acc: 0.8272
Epoch 287/300
891/891 [==============================] - 0s 35us/step - loss: 0.4150 - acc: 0.8283
Epoch 288/300
891/891 [==============================] - 0s 35us/step - loss: 0.4154 - acc: 0.8249
Epoch 289/300
891/891 [==============================] - 0s 35us/step - loss: 0.4138 - acc: 0.8283
Epoch 290/300
891/891 [==============================] - 0s 35us/step - loss: 0.4139 - acc: 0.8294
Epoch 291/300
891/891 [==============================] - 0s 35us/step - loss: 0.4150 - acc: 0.8272
Epoch 292/300
891/891 [==============================] - 0s 35us/step - loss: 0.4148 - acc: 0.8305
Epoch 293/300
891/891 [==============================] - 0s 35us/step - loss: 0.4151 - acc: 0.8272
Epoch 294/300
891/891 [==============================] - 0s 18us/step - loss: 0.4135 - acc: 0.8272
Epoch 295/300
891/891 [==============================] - 0s 35us/step - loss: 0.4152 - acc: 0.8283
Epoch 296/300
891/891 [==============================] - 0s 35us/step - loss: 0.4171 - acc: 0.8283
Epoch 297/300
891/891 [==============================] - 0s 35us/step - loss: 0.4164 - acc: 0.8215
Epoch 298/300
891/891 [==============================] - 0s 35us/step - loss: 0.4199 - acc: 0.8294
Epoch 299/300
891/891 [==============================] - 0s 35us/step - loss: 0.4152 - acc: 0.8272
Epoch 300/300
891/891 [==============================] - 0s 18us/step - loss: 0.4145 - acc: 0.8305
<keras.callbacks.History at 0x198f5d8d5f8>
y_pred_nn = np.round(model.predict(X_test))
y_pred_nn_list = []
for i in range(len(y_pred_nn)):
y_pred_nn_list.append(int(y_pred_nn[i][0]))nn_etc = accuracy_score(y_pred_nn,y_test)
print("nn acc: ",nn_etc)
predictions.to_csv("predictions.csv",index='PassengerId')nn acc: 0.9641148325358851