tf.ragged.map_flat_values

tf.ragged.map_flat_values 可以理解不規則矩陣的map

import tensorflow as tf 


rt = tf.ragged.constant([[1, 2, 3], [], [4, 5], [6]])

tf.ragged.map_flat_values(tf.ones_like, rt).to_list()
tf.ragged.map_flat_values(tf.multiply, rt, rt).to_list()

tf.ragged.map_flat_values 比下面更通用

import numpy as np

import tensorflow as tf
elems = np.array([1, 2, 3, 4, 5, 6])
squares = tf.map_fn(lambda x: x * x, elems)

print(squares)

 ids_map_layer = tf.keras.layers.Lambda(lambda x:tf.ragged.map_flat_values(ids_table.lookup, x))