在TensorFlow中实现矩阵维度扩展

一般TensorFlow中扩展维度可以使用tf.expand_dims()。近来发现另一种可以直接运用取数据操作符[]就能扩展维度的方法。

用法很简单,在要扩展的维度上加上tf.newaxis就行了。

foo = tf.constant([[1,2,3], [4,5,6], [7,8,9]])
print(foo[tf.newaxis, :, :].eval()) # => [[[1,2,3], [4,5,6], [7,8,9]]]
print(foo[:, tf.newaxis, :].eval()) # => [[[1,2,3]], [[4,5,6]], [[7,8,9]]]
print(foo[:, :, tf.newaxis].eval()) # => [[[1],[2],[3]], [[4],[5],[6]],[[7],[8],[9]]]

参考:

https://tensorflow.google.cn/api_docs/python/tf/Tensor?hl=en#__getitem__

补充知识:tensorflow 利用expand_dims和squeeze扩展和压缩tensor维度

在利用tensorflow进行文本挖掘工作的时候,经常涉及到维度扩展和压缩工作。比如对文本进行embedding操作完成之后,若要进行卷积操作,就需要对embedded的向量扩展维度,将[batch_size, embedding_dims]扩展成为[batch_size, embedding_dims, 1],利用tf.expand_dims(input, -1)就可实现,反过来用squeeze(input, -1)或者tf.squeeze(input)也可以把最第三维去掉。

tf.expand_dims()
tf.squeeze()

tf.expand_dims()

tf.expand_dims(input, axis=None, name=None, dim=None)

在第axis位置增加一个维度.

给定张量输入,此操作在输入形状的维度索引轴处插入1的尺寸。 尺寸索引轴从零开始; 如果您指定轴的负数,则从最后向后计数。

如果要将批量维度添加到单个元素,则此操作非常有用。 例如,如果您有一个单一的形状[height,width,channels],您可以使用expand_dims(image,0)使其成为1个图像,这将使形状[1,高度,宽度,通道]。

例子

# 't' is a tensor of shape [2]
shape(expand_dims(t, 0)) ==> [1, 2]
shape(expand_dims(t, 1)) ==> [2, 1]
shape(expand_dims(t, -1)) ==> [2, 1]
# 't2' is a tensor of shape [2, 3, 5]
shape(expand_dims(t2, 0)) ==> [1, 2, 3, 5]
shape(expand_dims(t2, 2)) ==> [2, 3, 1, 5]
shape(expand_dims(t2, 3)) ==> [2, 3, 5, 1]

tf.squeeze()

tf.squeeze(input, axis=None, name=None, squeeze_dims=None)

直接上例子

# 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
 shape(squeeze(t)) ==> [2, 3]
# 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
 shape(squeeze(t, [2, 4])) ==> [1, 2, 3, 1]

以上这篇在TensorFlow中实现矩阵维度扩展就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持我们。

(0)

相关推荐

  • pytorch实现Tensor变量之间的转换

    系统默认是torch.FloatTensor类型 data = torch.Tensor(2,3)是一个2*3的张量,类型为FloatTensor data.cuda()就转换为GPU的张量类型,torch.cuda.FloatTensor类型 (1) CPU或GPU之间的张量转换 在Tensor后加long(), int(), double(),float(),byte()等函数就能将Tensor进行类型转换type()函数, data为Tensor数据类型,data.type()为给出dat

  • 在keras 中获取张量 tensor 的维度大小实例

    在进行keras 网络计算时,有时候需要获取输入张量的维度来定义自己的层.但是由于keras是一个封闭的接口.因此在调用由于是张量不能直接用numpy 里的A.shape().这样的形式来获取.这里需要调用一下keras 作为后端的方式来获取.当我们想要操作时第一时间就想到直接用 shape ()函数.其实keras 中真的有shape()这个函数. shape(x)返回一个张量的符号shape,符号shape的意思是返回值本身也是一个tensor, 示例: >>> from keras

  • PyTorch中Tensor的维度变换实现

    对于 PyTorch 的基本数据对象 Tensor (张量),在处理问题时,需要经常改变数据的维度,以便于后期的计算和进一步处理,本文旨在列举一些维度变换的方法并举例,方便大家查看. 维度查看:torch.Tensor.size() 查看当前 tensor 的维度 举个例子: >>> import torch >>> a = torch.Tensor([[[1, 2], [3, 4], [5, 6]]]) >>> a.size() torch.Size

  • PyTorch中Tensor的拼接与拆分的实现

    拼接张量:torch.cat() .torch.stack() torch.cat(inputs, dimension=0) → Tensor 在给定维度上对输入的张量序列 seq 进行连接操作 举个例子: >>> import torch >>> x = torch.randn(2, 3) >>> x tensor([[-0.1997, -0.6900, 0.7039], [ 0.0268, -1.0140, -2.9764]]) >>&

  • 在TensorFlow中实现矩阵维度扩展

    一般TensorFlow中扩展维度可以使用tf.expand_dims().近来发现另一种可以直接运用取数据操作符[]就能扩展维度的方法. 用法很简单,在要扩展的维度上加上tf.newaxis就行了. foo = tf.constant([[1,2,3], [4,5,6], [7,8,9]]) print(foo[tf.newaxis, :, :].eval()) # => [[[1,2,3], [4,5,6], [7,8,9]]] print(foo[:, tf.newaxis, :].eva

  • 浅谈tensorflow中Dataset图片的批量读取及维度的操作详解

    三维的读取图片(w, h, c): import tensorflow as tf import glob import os def _parse_function(filename): # print(filename) image_string = tf.read_file(filename) image_decoded = tf.image.decode_image(image_string) # (375, 500, 3) image_resized = tf.image.resize

  • tensorflow 利用expand_dims和squeeze扩展和压缩tensor维度方式

    在利用tensorflow进行文本挖掘工作的时候,经常涉及到维度扩展和压缩工作.比如对文本进行embedding操作完成之后,若要进行卷积操作,就需要对embedded的向量扩展维度,将[batch_size, embedding_dims]扩展成为[batch_size, embedding_dims, 1],利用tf.expand_dims(input, -1)就可实现,反过来用squeeze(input, -1)或者tf.squeeze(input)也可以把最第三维去掉. tf.expan

  • numpy和tensorflow中的各种乘法(点乘和矩阵乘)

    点乘和矩阵乘的区别: 1)点乘(即" * ") ---- 各个矩阵对应元素做乘法 若 w 为 m*1 的矩阵,x 为 m*n 的矩阵,那么通过点乘结果就会得到一个 m*n 的矩阵. 若 w 为 m*n 的矩阵,x 为 m*n 的矩阵,那么通过点乘结果就会得到一个 m*n 的矩阵. w的列数只能为 1 或 与x的列数相等(即n),w的行数与x的行数相等 才能进行乘法运算. 2)矩阵乘 ---- 按照矩阵乘法规则做运算 若 w 为 m*p 的矩阵,x 为 p*n 的矩阵,那么通过矩阵相乘结

  • Pytorch中expand()的使用(扩展某个维度)

    目录 Pytorch expand()的使用 Pytorch expand()函数 返回tensor的一个新视图 note:使用expand()函数的时候 Pytorch expand()的使用 有两点需要注意,无论是 expand() 还是 expand_as(): 1.只能在第0维扩展一个维数,比如原来是是(1,3,4)==>(2,1,3,4),而在其他维度扩展不可以(1,3,4)==>(1,2,3,4)[错误] 2.如果不增加维数,只是增加维度,要增加的原维度必须是1才可以在该维度增加维

  • 对Tensorflow中权值和feature map的可视化详解

    前言 Tensorflow中可以使用tensorboard这个强大的工具对计算图.loss.网络参数等进行可视化.本文并不涉及对tensorboard使用的介绍,而是旨在说明如何通过代码对网络权值和feature map做更灵活的处理.显示和存储.本文的相关代码主要参考了github上的一个小项目,但是对其进行了改进. 原项目地址为(https://github.com/grishasergei/conviz). 本文将从以下两个方面进行介绍: 卷积知识补充 网络权值和feature map的可

  • Python3 Tensorlfow:增加或者减小矩阵维度的实现

    1.增加维度 下面给出两个样例 样例1: [1, 2, 3] ==> [[1],[2],[3]] import tensorflow as tf a = tf.constant([1, 2, 3]) b = tf.expand_dims(a,1) with tf.Session() as sess: a_, b_ = sess.run([a, b]) print('a:') print(a_) print('b:') print(b_) 输出结果 a: [1 2 3] b: [[1] [2] [

  • TensorFlow中tf.batch_matmul()的用法

    TensorFlow中tf.batch_matmul()用法 如果有两个三阶张量,size分别为 a.shape = [100, 3, 4] b.shape = [100, 4, 5] c = tf.batch_matmul(a, b) 则c.shape = [100, 3, 5] //将每一对 3x4 的矩阵与 4x5 的矩阵分别相乘.batch_size不变 100为张量的batch_size.剩下的两个维度为数据的维度. 不过新版的tensorflow已经移除了上面的函数,使用时换为tf.

  • 对Tensorflow中的矩阵运算函数详解

    tf.diag(diagonal,name=None) #生成对角矩阵 import tensorflowas tf; diagonal=[1,1,1,1] with tf.Session() as sess: print(sess.run(tf.diag(diagonal))) #输出的结果为[[1 0 0 0] [0 1 0 0] [0 0 1 0] [0 0 0 1]] tf.diag_part(input,name=None) #功能与tf.diag函数相反,返回对角阵的对角元素 imp

随机推荐