tf.keras.layers.Embedding 嵌入层 示例

import tensorflow as tf
import numpy as np

将正整数（索引）转换为固定大小的密集向量。此层只能用作模型中的第一层

embedding = tf.keras.layers.Embedding(1000, 64, input_length=10)
print(embedding)

<tensorflow.python.keras.layers.embeddings.Embedding object at 0x000001F3B4355EB0>

input_array = np.random.randint(1000, size=(32, 10))
print(input_array.shape)

(32, 10)

output_array = embedding(input_array)
print(output_array.shape)

(32, 10, 64)