TensorFlow tf.keras.layers.GlobalAveragePooling1D

参数

steps 是时间序列的意思,就是一句话包含了多少个词,features是每个词的特征,可以联想图片处理中一个像素点对应3个通道(channel,就是feathures,其对应的就是filters)

参数	描述
data_format	channels_last (default):(batch, steps, features);channels_first:(batch, features, steps)

输入形状

三维的张量
channels_last:(batch_size, steps, features)
channels_first:(batch_size, features, steps)

输出形状

二维张量:(batch_size, features)

应用:

from tensorflow import keras
import numpy as np

data = np.array([[0,0,0],[1,1,1]])
emb = keras.layers.Embedding(input_dim=2, output_dim=3, input_length=3)
emb(data)