from keras.layers import Dense, Dropout, LSTM, Embedding
from keras.preprocessing.sequence import pad_sequences
from keras.models import Sequential
import pandas as pd
import numpy as np
def read_data():
min_len = -1
max_len = -1
X_train = []
Y_train = []
X_test = []
Y_test = []
for i in range(1, 7):
loc = 'without syndrome/control' + str(i) + '.xlsx'
excel_data = pd.read_excel(loc)
data = excel_data['HR'].tolist()
if min_len == -1 or min_len > len(data):
min_len = len(data)
if max_len == -1 or max_len < len(data):
max_len = len(data)
X_train.append(data)
Y_train.append(0)
for i in range(1, 7):
loc = 'with syndrome/NAS' + str(i) + '.xlsx'
excel_data = pd.read_excel(loc)
data = excel_data['HR'].tolist()
if min_len == -1 or min_len > len(data):
min_len = len(data)
if max_len == -1 or max_len < len(data):
max_len = len(data)
X_train.append(np.array(data))
Y_train.append(1)
for i in range(1, 3):
loc = 'test/test' + str(i) + '.xlsx'
excel_data = pd.read_excel(loc)
data = excel_data['HR'].tolist()
if min_len == -1 or min_len > len(data):
min_len = len(data)
if max_len == -1 or max_len < len(data):
max_len = len(data)
X_test.append(data)
Y_test.append(i)
truncated = []
for xt in X_train:
truncated.append(xt[:min_len]);
truncated_test = []
for xt in X_test:
truncated_test.append(xt[:min_len]);
return min_len, np.array(truncated), np.array(Y_train), np.array(truncated_test), np.array(Y_test)
def create_model(input_length):
print('Creating model...')
model = Sequential()
model.add(Embedding(input_dim=300, output_dim=50, input_length = input_length))
model.add(LSTM(256, activation='sigmoid', return_sequences=True))
model.add(Dropout(0.5))
model.add(LSTM(256, activation='sigmoid'))
model.add(Dropout(0.5))
model.add(Dense(1, activation='sigmoid'))
print('Compiling...')
model.compile(loss='binary_crossentropy', optimizer='rmsprop', metrics=['accuracy'])
return model
if __name__ == '__main__':
min_len, X_train, Y_train, X_test, Y_test = read_data()
print(min_len)
model = create_model(min_len)
print('Fitting model...')
hist = model.fit(X_train, Y_train, batch_size=64, epochs=10, validation_split=0.1, verbose=1)
score, acc = model.evaluate(X_test, Y_test, batch_size=1)
print('Test score:', score)
print('Test accuracy:', acc)
