1.添加网络结构
model.add(...)
model.add(...)
...
2.模型编译
model.compile(loss='...',omeptimizer='...',metrics=['accuracy'])
3.模型训练
history = model.fit(x_train,y_train,batchsize=64,epoch=epochs,validation=(x_test,y_test))
4.打印结果
plt.plot(history.epoch, history.history['loss'], label='train loss+error')
plt.plot(history.epoch, history.history['val_loss'], label='val_error')
plt.plot(history.epoch, history.history['acc'],label='train_acc')
plt.plot(history.epoch, history.history['val_acc'], label = 'val_acc')
plt.legend()