keras多输出多输出示例（keras教程一）

参考 keras官网

问题描述：通过模型对故障单按照优先级排序并制定给正确的部门。

输入：

• 票证的标题（文本输入），
• 票证的文本正文（文本输入），以及
• 用户添加的任何标签（分类输入）

输出：

• 优先级分数介于0和1之间（sigmoid 输出），以及
• 应该处理票证的部门（部门范围内的softmax输出）

import keras
import numpy as np

num_tags = 12  # Number of unique issue tags
num_words = 10000  # 预处理文本数据时获得的词汇量
num_departments = 4  # Number of departments for predictions

title_input = keras.Input(
    shape=(None,), name="title"
)  # Variable-length sequence of ints
body_input = keras.Input(shape=(None,), name="body")  # Variable-length sequence of ints
tags_input = keras.Input(
    shape=(num_tags,), name="tags"
)  # Binary vectors of size `num_tags`

# Embed each word in the title into a 64-dimensional vector
title_features = keras.layers.Embedding(num_words, 64)(title_input)
# Embed each word in the text into a 64-dimensional vector
body_features = keras.layers.Embedding(num_words, 64)(body_input)

# Reduce sequence of embedded words in the title into a single 128-dimensional vector
title_features = keras.layers.LSTM(128)(title_features)
# Reduce sequence of embedded words in the body into a single 32-dimensional vector
body_features = keras.layers.LSTM(32)(body_features)

# Merge all available features into a single large vector via concatenation
x = keras.layers.concatenate([title_features, body_features, tags_input])

# Stick a logistic regression for priority prediction on top of the features
priority_pred = keras.layers.Dense(1, name="priority")(x)
# Stick a department classifier on top of the features
department_pred = keras.layers.Dense(num_departments, name="department")(x)

# Instantiate an end-to-end model predicting both priority and department
model = keras.Model(
    inputs=[title_input, body_input, tags_input],
    outputs=[priority_pred, department_pred],
)
model.summary()
keras.utils.plot_model(model, "multi_input_and_output_model.png", show_shapes=True)

# model.compile(
#     optimizer=keras.optimizers.RMSprop(1e-3),
#     loss={
#         "priority": "binary_crossentropy",
#         "department": "categorical_crossentropy",
#     },
#     loss_weights=[1.0, 0.2],
# )

model.compile(
    optimizer=keras.optimizers.RMSprop(1e-3),
    loss={
        "priority": "binary_crossentropy",
        "department": "categorical_crossentropy",
    },
    loss_weights={'priority': 1., 'department': 0.2},)
# Dummy input data
title_data = np.random.randint(num_words, size=(1280, 10))
body_data = np.random.randint(num_words, size=(1280, 100))
tags_data = np.random.randint(2, size=(1280, num_tags)).astype("float32")

# Dummy target data
priority_targets = np.random.random(size=(1280, 1))
dept_targets = np.random.randint(2, size=(1280, num_departments))

model.fit(
    {"title": title_data, "body": body_data, "tags": tags_data},
    {"priority": priority_targets, "department": dept_targets},
    epochs=2,
    batch_size=32,
)
model.save("path_to_my_model")
model = keras.models.load_model("path_to_my_model")

环境：keras==2.2.4 tensorflow==1.12.0

模型结构

模型参数

__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to 
==================================================================================================
title (InputLayer) (None, None) 0 
__________________________________________________________________________________________________
body (InputLayer) (None, None) 0 
__________________________________________________________________________________________________
embedding_1 (Embedding) (None, None, 64) 640000 title[0][0] 
__________________________________________________________________________________________________
embedding_2 (Embedding) (None, None, 64) 640000 body[0][0] 
__________________________________________________________________________________________________
lstm_1 (LSTM) (None, 128) 98816 embedding_1[0][0] 
__________________________________________________________________________________________________
lstm_2 (LSTM) (None, 32) 12416 embedding_2[0][0] 
__________________________________________________________________________________________________
tags (InputLayer) (None, 12) 0 
__________________________________________________________________________________________________
concatenate_1 (Concatenate) (None, 172) 0 lstm_1[0][0] 
lstm_2[0][0] 
tags[0][0] 
__________________________________________________________________________________________________
priority (Dense) (None, 1) 173 concatenate_1[0][0] 
__________________________________________________________________________________________________
department (Dense) (None, 4) 692 concatenate_1[0][0] 
==================================================================================================
Total params: 1,392,097
Trainable params: 1,392,097

　　

参数计算方法

该模型有前馈神经网络和LSTM。参考深度学习模型参数计算。