return model
model = build_model()
EPOCHS = 1000
early_stop = keras.callbacks.EarlyStopping(monitor='val_loss', patience=100)
history = model.fit(X_train, y_train,
epochs=EPOCHS, validation_split = 0.2, verbose=0,
callbacks=[early_stop,tfdocs.modeling.EpochDots()])
hist = pd.DataFrame(history.history)
hist['epoch'] = history.epoch
hist.tail()
loss, mae, mse = model.evaluate(X_test,y_test)
test_predictions = model.predict(X_test)
NN_score = r2_score(y_test,test_predictions)
print(NN_score)
plotter = tfdocs.plots.HistoryPlotter(smoothing_std=2)
plotter.plot({'Basic': history}, metric = "mae")
plt.ylabel('MAE ')
plt.savefig("MAE_loss_with_early.png")
plotter.plot({'Basic': history}, metric = "mse")
plt.ylabel('MSE')
plt.savefig('MSE_loss_with_early.png')
X_train, X_test, y_train, y_test = train_test_split(X_scaled,
y,
test_size=0.3,
random_state=42)
def model():
model = Sequential()
model.add(
Dense(256,
activation='relu',
kernel_initializer='he_normal',
input_dim=13))
model.add(Dropout(0.2))
model.add(Dense(128, activation='relu', kernel_initializer='he_normal'))
model.add(Dropout(0.3))
model.add(Dense(16, activation='relu', kernel_initializer='he_normal'))
model.add(Dense(1))
model.compile(loss='mse', optimizer='adam')
return model
model = model()
model.fit(X_train, y_train, shuffle=True, epochs=1000, verbose=0)
print('RMSE : {:.2f}'.format(np.sqrt(model.evaluate(X_test, y_test))))
y_pred = model.predict(X_test)
print(f'Accuracy: {r2_score(y_test, y_pred)*100:.2f}%')
