Neural network trained in kaggles lower back pain dataset

kaggle_lower_back_pain.py

#This is a fun project for kaggles Lower Back Pain dataset
#https://www.kaggle.com/sammy123/lower-back-pain-symptoms-dataset
#Used a simple multi layer perceptron to train
#contact: https://gananath.github.io

import pandas as pd
import numpy as np
from keras.models import Sequential
from keras.layers import Dense,Dropout
from sklearn.cross_validation import train_test_split
from keras.utils import np_utils
#seed for random number reproduction
np.random.seed(2016)

#import dataset
data = pd.read_csv("/home/gananath/Desktop/Dataset_spine.csv",header=0)
data.head()

#mapping the data
d = {'Abnormal' : 1, 'Normal' : 0}
data['Class_att'] = data['Class_att'].map(d)

data = data.rename(columns={'Col1' : 'pelvic_incidence','Col2' : 'pelvic_tilt','Col3' : 'lumbar_lordosis_angle',
                           'Col4' : 'sacral_slope','Col5': 'pelvic_radius','Col6' : 'degree_spondylolisthesis',
                           'Col7' : 'pelvic_slope','Col8' : 'Direct_tilt','Col9': 'thoracic_slope',
                           'Col10' : 'cervical_tilt','Col11' : 'sacrum_angle','Col12' : 'scoliosis_slope'})
del data['Unnamed: 13']
data=data.reindex(np.random.permutation(data.index))
data.head()

features = list(data.columns[:12])
features

X = data[features]
y = data["Class_att"]
# one hot encode outputs
y=np_utils.to_categorical(y)


X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=1)

#model building
model = Sequential()
model.add(Dense(X_train.shape[1], input_dim=X_train.shape[1], init='normal', activation='relu'))
model.add(Dense(20, init='normal', activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(2,init='normal', activation='softmax'))
#compiling model
model.compile(loss='categorical_crossentropy', optimizer='adam',metrics=['accuracy'])

model.summary()

history=model.fit(X_train.as_matrix(), y_train,validation_data=(X_test.as_matrix(),y_test), shuffle=True, nb_epoch=100, batch_size=15)

print "Accuracy: " + str(np.mean(history.history['val_acc']))