Keras 모델 생성/학습 - 당뇨병 예측 모델

데이터 처리

# 1. Pandas 가져오기
import pandas as pd

# 2. 데이터 불러오기
data = pd.read_csv('diabetes_data.csv')
data.head()

# 데이터 통계치 확인하기

data.describe()

# 3. X/y 나누기
X = data.iloc[:,:-1].values
y = data.iloc[:,-1].values

# values를 붙여서 numpy array형태로 가져온다!!!!! 매우 중요하다!!!! 큰 깨달음
print(X.shape)
print(y.shape)

# 4. Train set, Test set 나누기
from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.25, random_state=42)


X_val, X_test, y_val, y_test = train_test_split(X_test,y_test, 
                                                test_size=0.5,
                                                random_state=19)
print(X_train.shape)
print(X_val.shape)
print(X_test.shape)
print(y_train.shape)
print(y_val.shape)
print(y_test.shape)

---

Keras 모델 만들기

# 5. Keras 패키지 가져오기

from keras.models import Sequential
from keras.layers import Dense

# 6. MLP 모델 생성

model = Sequential()

model.add(Dense(units=64, input_dim=8, activation='relu'))
model.add(Dense(units=128, activation='relu'))
model.add(Dense(units=1, activation='sigmoid'))

model.summary()

# 7. Compile - Optimizer, Loss function 설정
model.compile(loss='binary_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])

# 8. 학습시키기
from keras.callbacks import EarlyStopping, ModelCheckpoint

early_stop = EarlyStopping(patience=20)
batch_size = 32
epochs = 200
history = model.fit(X_train,y_train,
                  batch_size=batch_size,
                  nb_epoch=epochs,
                  validation_data=(X_val,y_val),
                  callbacks=[early_stop],
                  verbose=1
                  )

# 9. 모델 평가하기
train_acc = model.evaluate(X_train,y_train)
test_acc = model.evaluate(X_test,y_test)

print(train_acc)
print(test_acc)

시각화

# 10. 학습 시각화하기
import matplotlib.pyplot as plt

plt.plot(history.history['accuracy'])
plt.plot(history.history['val_accuracy'])

plt.title('Accuracy')
plt.xlabel('epoch')
plt.ylabel('accuracy')
plt.legend(['train','val'], loc = 'upper left')
plt.show()

# loss 값 그래프 그리기
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])

plt.title('loss')
plt.xlabel('epoch')
plt.ylabel('loss')
plt.legend(['train','val'], loc = 'upper left')
plt.show()

# 11. 모델 저장
model_path ='diabet_model.h5'
model.save(model_path)

# 12. 모델 불러오기

from keras.models import load_model

loaded_model = load_model(model_path)
print(loaded_model.summary())