tf.gather和tf.gather_nd

最新推荐文章于 2024-01-17 08:00:00 发布

ICDAT_GO

最新推荐文章于 2024-01-17 08:00:00 发布

阅读量887

点赞数 1

分类专栏： TensorFlow学习笔记文章标签： tensorflow

本文链接：https://blog.csdn.net/qq_40657582/article/details/121485368

版权

TensorFlow学习笔记专栏收录该内容

1 篇文章 0 订阅

订阅专栏

tf.gather

Gather slices from params axis according to indices.

tf.gather(params, indices, validate_indices=None, axis=None, batch_dims=0, name=None)

Example：

# data: [classes, students, subjects]
data = tf.ones([4,35,8])
print(data.shape)    # TensorShape([4, 35, 8])

# sample several classes
data = tf.gather(data, axis=0, indices=[2,3])
print(data.shape)    # TensorShape([2, 35, 8])

tf.gather_nd

Gather slices from params into a Tensor with shape specified by indices.

tf.gather_nd(params, indices, batch_dims=0, name=None)

Example：

# data: [classes, students, subjects]
data = tf.ones([4,35,8])
print(data.shape)    # TensorShape([4, 35, 8])

# sample several (classes and students)
# for instance: [class1_student1, class2_studnet2, class3_student3, class4_student4]
data = tf.gather_nd(data, [[0,0],[1,1],[2,2],[3,3]])
print(data.shape)    # TensorShape([4, 8])