睿智的keras深度学习（二）——函数式API构建顺序模型快速上手

系类文章往期回顾

睿智的keras深度学习（零）——keras使用时可能遇到的参数及含义（持续更新
睿智的keras深度学习（一）——阈值逻辑单元和多层感知机

前言

本篇文章只是用来说明模型的构架和数据的传入，并不代表所使用的数据集就应该按照文章的方法进行训练，读者仅需掌握API的使用即可，不必过度注重数据集和模型训练

加载数据集

这里使用fetch_california_housing数据集进行操作

from sklearn.datasets import fetch_california_housing
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler

housing = fetch_california_housing()
X_train_full, X_test, y_train_full, y_test = train_test_split(housing.data, housing.target)
X_train, X_valid, y_train, y_valid = train_test_split(X_train_full, y_train_full)

数据归一化

为了提高模型的可靠性我们很多时候都需要进行数据的归一化

scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_valid = scaler.transform(X_valid)
X_test = scaler.transform(X_test)

使用函数式API构建残差模型

构建模型

from keras.layers import Input, Concatenate
from keras import Model

input_ = Input(shape=X_train.shape[1:])
hidden1 = Dense(30, activation="relu")(input_)
hidden2 = Dense(30, activation="relu")(hidden1)
concat = Concatenate()([input_, hidden2])
output = Dense(1)(concat)
model = Model(inputs=[input_], outputs=[output])

plot_model(model)

使用函数式API构建多输入模型

构建模型

input_A = Input(shape=[5], name="wide_input")
input_B = Input(shape=[6], name="deep_input")
hidden1 = Dense(30, activation="relu")(input_B)
hidden2 = Dense(30, activation="relu")(hidden1)
concat = Concatenate()([input_A, hidden2])
output = Dense(1, name="output")(concat)
model = Model(inputs=[input_A, input_B], outputs=[output])

plot_model(model)

模型训练

from tensorflow.keras.optimizers import SGD

model.compile(loss="mse", optimizer=SGD(lr=1e-3))
X_train_A, X_train_B = X_train[:, :5], X_train[:, 2:]
X_valid_A, X_valid_B = X_valid[:, :5], X_valid[:, 2:]
X_test_A, X_test_B = X_test[:, :5], X_test[:, 2:]

X_new_A, X_new_B = X_test_A[:3], X_test_B[:3]

history = model.fit((X_train_A, X_train_B), y_train, epochs=20, validation_data=((X_valid_A, X_valid_B), y_valid))

mse_test = model.evaluate((X_test_A, X_test_B), y_test)
y_pred = model.predict((X_new_A, X_new_B))
y_pred

使用函数式API构建多输入多输出模型

构建模型

output = Dense(1, name="main_output")(concat)
aux_output = Dense(1, name="aux_output")(hidden2)
model = Model(inputs=[input_A, input_B], outputs=[output, aux_output])

plot_model(model)

模型训练和评估

由于模型是多输出模型，所以指定两个损失函数，并且可以给他们分配不同的权值，权值的大小会影响模型的最终结果

model.compile(loss=["mse", "mse"], loss_weights=[0.9, 0.1], optimizer="sgd")
history = model.fit([X_train_A, X_train_B], [y_train, y_train], epochs=20, validation_data=([X_valid_A, X_valid_B], [y_valid, y_valid]))

total_loss, main_loss, aux_loss = model.evaluate([X_test_A, X_test_B], [y_test, y_test])
y_pred_main, y_pred_aux = model.predict([X_new_A, X_new_B])

total_loss, main_loss, aux_loss

y_pred_main, y_pred_aux