深度学习框架Tensor张量的操作使用

- 重点掌握基本张量使用及与numpy的区别

- 掌握张量维度操作（拼接、维度扩展、压缩、转置、重复……）

numpy基本操作:

• numpy学习4：NumPy基本操作

• NumPy 教程

1. Tensorflow

• Tensorflow一些常用基本概念与函数（1,2,3,4）

• tensorflow与numpy函数的选择

Tensorflow 和numpy区别

相同点： 都提供n位数组
不同点： numpy支持ndarray，而Tensorflow里有tensor；numpy不提供创建张量函数和求导，也不提供GPU支持。

-tensorflow中tf.random_normal和tf.truncated_normal的区别

代码
a = tf.Variable(tf.random_normal([,],seed=))
b = tf.Variable(tf.truncated_normal([,],seed=))
init = tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(init)
    print(sess.run(a))
    print(sess.run(b))

输出：
[[-0.81131822  1.48459876]
 [ 0.06532937 -2.44270396]]
[[-0.85811085 -0.19662298]
 [ 0.13895047 -1.22127688]]

拼接函数：tf.concat()等价于torch.cat()

转置函数：torch中：对二维Tensor转置操作transpose(dim1,dim2)或者直接t();

对多维Tensor转置操作permute(dim1,dim2,...,dimn),

增加维度：TensorFlow中，想要维度增加一维，可以使用tf.expand_dims(input,dim,name=None)函数

Torch中，使用nn.unsqueeze(pos [,numInputDims])在pos位置上插入1.

• 『TensorFlow』张量拼接_调整维度_切片

• TensorFlow 张量相关操作

• TensorFlow进阶（二）---张量的操作

2. Pytorch

import torch
import numpy as np

np_data = np.arange().reshape((, ))

# numpy 转为 pytorch格式

torch_data = torch.from_numpy(np_data)
print(
  '\n numpy', np_data,
  '\n torch', torch_data,
)
'''
 numpy [[  ]
 [  ]]
 torch 

[torch.LongTensor of size 2x3]
'''
# torch 转为numpy
tensor2array = torch_data.numpy()
print(tensor2array)
"""
[[  ]
 [  ]]
"""
# 运算符
# abs 、 add 、和numpy类似
data = [[, ], [, ]]
tensor = torch.FloatTensor(data)    # 转为32位浮点数，torch接受的都是Tensor的形式，所以运算前先转化为Tensor
print(
  '\n numpy', np.matmul(data, data),
  '\n torch', torch.mm(tensor, tensor)    # torch.dot()是点乘
)
'''
 numpy [[  ]
 [ ]]
 torch 

[torch.FloatTensor of size 2x2]
'''

- pytorch0.3: x = Variable(torch.rand(5, 3, 224, 224), requires_grad=True).cuda()

# -*- coding: utf- -*-
"""
Neural Networks
https://pytorch.org/tutorials/beginner/blitz/neural_networks_tutorial.html#sphx-glr-beginner-blitz-neural-networks-tutorial-py
===============

Neural networks can be constructed using the ``torch.nn`` package.

Now that you had a glimpse of ``autograd``, ``nn`` depends on
``autograd`` to define models and differentiate them.
An ``nn.Module`` contains layers, and a method ``forward(input)``\ that
returns the ``output``.

For example, look at this network that classifies digit images:

.. figure:: /_static/img/mnist.png
   :alt: convnet

   convnet

It is a simple feed-forward network. It takes the input, feeds it
through several layers one after the other, and then finally gives the
output.

A typical training procedure for a neural network is as follows:

- Define the neural network that has some learnable parameters (or
  weights)
- Iterate over a dataset of inputs
- Process input through the network
- Compute the loss (how far is the output from being correct)
- Propagate gradients back into the network’s parameters
- Update the weights of the network, typically using a simple update rule:
  ``weight = weight - learning_rate * gradient``

Define the network
------------------

Let’s define this network:
"""
import torch
import torch.nn as nn
import torch.nn.functional as F

class Net(nn.Module):

    def __init__(self):
        super(Net, self).__init__()
        #  input image channel,  output channels, 5x5 square convolution
        # kernel
        self.conv1 = nn.Conv2d(, , )
        self.conv2 = nn.Conv2d(, , )
        # an affine operation: y = Wx + b
        self.fc1 = nn.Linear( *  * , )
        self.fc2 = nn.Linear(, )
        self.fc3 = nn.Linear(, )

    def forward(self, x):
        # Max pooling over a (, ) window
        x = F.max_pool2d(F.relu(self.conv1(x)), (, ))
        # If the size is a square you can only specify a single number
        x = F.max_pool2d(F.relu(self.conv2(x)), )
        x = x.view(-, self.num_flat_features(x))
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x

    def num_flat_features(self, x):
        size = x.size()[:]  # all dimensions except the batch dimension
        num_features =
        for s in size:
            num_features *= s
        return num_features

net = Net()
print(net)

########################################################################
# You just have to define the ``forward`` function, and the ``backward``
# function (where gradients are computed) is automatically defined for you
# using ``autograd``.
# You can use any of the Tensor operations in the ``forward`` function.
#
# The learnable parameters of a model are returned by ``net.parameters()``

params = list(net.parameters())
print(len(params))
print(params[].size())  # conv1's .weight

########################################################################
# Let try a random 32x32 input.
# Note: expected input size of this net (LeNet) is 32x32. To use this net on
# MNIST dataset, please resize the images from the dataset to 32x32.

input = torch.randn(, , , )
out = net(input)
print(out)

########################################################################
# Zero the gradient buffers of all parameters and backprops with random
# gradients:
net.zero_grad()
out.backward(torch.randn(, ))

########################################################################
# .. note::
#
#     ``torch.nn`` only supports mini-batches. The entire ``torch.nn``
#     package only supports inputs that are a mini-batch of samples, and not
#     a single sample.
#
#     For example, ``nn.Conv2d`` will take in a 4D Tensor of
#     ``nSamples x nChannels x Height x Width``.
#
#     If you have a single sample, just use ``input.unsqueeze()`` to add
#     a fake batch dimension.
#
# Before proceeding further, let's recap all the classes you’ve seen so far.
#
# **Recap:**
#   -  ``torch.Tensor`` - A *multi-dimensional array* with support for autograd
#      operations like ``backward()``. Also *holds the gradient* w.r.t. the
#      tensor.
#   -  ``nn.Module`` - Neural network module. *Convenient way of
#      encapsulating parameters*, with helpers for moving them to GPU,
#      exporting, loading, etc.
#   -  ``nn.Parameter`` - A kind of Tensor, that is *automatically
#      registered as a parameter when assigned as an attribute to a*
#      ``Module``.
#   -  ``autograd.Function`` - Implements *forward and backward definitions
#      of an autograd operation*. Every ``Tensor`` operation creates at
#      least a single ``Function`` node that connects to functions that
#      created a ``Tensor`` and *encodes its history*.
#
# **At this point, we covered:**
#   -  Defining a neural network
#   -  Processing inputs and calling backward
#
# **Still Left:**
#   -  Computing the loss
#   -  Updating the weights of the network
#
# Loss Function
# -------------
# A loss function takes the (output, target) pair of inputs, and computes a
# value that estimates how far away the output is from the target.
#
# There are several different
# `loss functions <https://pytorch.org/docs/nn.html#loss-functions>`_ under the
# nn package .
# A simple loss is: ``nn.MSELoss`` which computes the mean-squared error
# between the input and the target.
#
# For example:

output = net(input)
target = torch.randn()  # a dummy target, for example
target = target.view(, -)  # make it the same shape as output
criterion = nn.MSELoss()

loss = criterion(output, target)
print(loss)

########################################################################
# Now, if you follow ``loss`` in the backward direction, using its
# ``.grad_fn`` attribute, you will see a graph of computations that looks
# like this:
#
# ::
#
#     input -> conv2d -> relu -> maxpool2d -> conv2d -> relu -> maxpool2d
#           -> view -> linear -> relu -> linear -> relu -> linear
#           -> MSELoss
#           -> loss
#
# So, when we call ``loss.backward()``, the whole graph is differentiated
# w.r.t. the loss, and all Tensors in the graph that has ``requires_grad=True``
# will have their ``.grad`` Tensor accumulated with the gradient.
#
# For illustration, let us follow a few steps backward:

print(loss.grad_fn)  # MSELoss
print(loss.grad_fn.next_functions[][])  # Linear
print(loss.grad_fn.next_functions[][].next_functions[][])  # ReLU

########################################################################
# Backprop
# --------
# To backpropagate the error all we have to do is to ``loss.backward()``.
# You need to clear the existing gradients though, else gradients will be
# accumulated to existing gradients.
#
#
# Now we shall call ``loss.backward()``, and have a look at conv1's bias
# gradients before and after the backward.

net.zero_grad()     # zeroes the gradient buffers of all parameters

print('conv1.bias.grad before backward')
print(net.conv1.bias.grad)

loss.backward()

print('conv1.bias.grad after backward')
print(net.conv1.bias.grad)

########################################################################
# Now, we have seen how to use loss functions.
#
# **Read Later:**
#
#   The neural network package contains various modules and loss functions
#   that form the building blocks of deep neural networks. A full list with
#   documentation is `here <https://pytorch.org/docs/nn>`_.
#
# **The only thing left to learn is:**
#
#   - Updating the weights of the network
#
# Update the weights
# ------------------
# The simplest update rule used in practice is the Stochastic Gradient
# Descent (SGD):
#
#      ``weight = weight - learning_rate * gradient``
#
# We can implement this using simple python code:
#
# .. code:: python
#
#     learning_rate = 0.01
#     for f in net.parameters():
#         f.data.sub_(f.grad.data * learning_rate)
#
# However, as you use neural networks, you want to use various different
# update rules such as SGD, Nesterov-SGD, Adam, RMSProp, etc.
# To enable this, we built a small package: ``torch.optim`` that
# implements all these methods. Using it is very simple:

import torch.optim as optim

# create your optimizer
optimizer = optim.SGD(net.parameters(), lr=0.01)

# in your training loop:
optimizer.zero_grad()   # zero the gradient buffers
output = net(input)
loss = criterion(output, target)
loss.backward()
optimizer.step()    # Does the update

###############################################################
# .. Note::
#
#       Observe how gradient buffers had to be manually set to zero using
#       ``optimizer.zero_grad()``. This is because gradients are accumulated
#       as explained in `Backprop`_ section.

原

• PyTorch 常用方法总结4：张量维度操作（拼接、维度扩展、压缩、转置、重复……）

• PyTorch中view的用法

我们会经常摆弄数据的维度，有时候是扩展（cat,expand），有时候又要压缩裁剪（squeeze,view），所以这些操纵向量维度的方法就尤其重要了，我把这些方法总结到这里，以供自己和朋友们参考。

涉及到的方法有：

torch.cat() torch.Tensor.expand()

torch.squeeze() torch.Tensor.repeat()

torch.Tensor.narrow() torch.Tensor.view()

torch.Tensor.resize_() torch.Tensor.permute()

另前三期总结的传送门：

4钟生成随机数Tensor的方法总结

Tensor常用数学运算

Tensor比大小

3.MXNet

-   MXNet for PyTorch users in 10 minutes : https://beta.mxnet.io/guide/to-mxnet/pytorch.html

- 一种原生API,一种Gluon API

- 原生API mnist实现: https://mxnet.incubator.apache.org/versions/master/tutorials/python/mnist.html

numpy和tensor的转换

tensorlfow
numpy转tensor
a = np.zeros((, ))
ta = tf.convert_to_tensor(a)
with tf.Session() as sess:
    print(sess.run(ta))

tensor转numpy
import tensorflow as tf
img1 = tf.constant(value=[[[[],[],[],[]],[[],[],[],[]],[[],[],[],[]],[[],[],[],[]]]],dtype=tf.float32)
img2 = tf.constant(value=[[[[],[],[],[]],[[],[],[],[]],[[],[],[],[]],[[],[],[],[]]]],dtype=tf.float32)
img = tf.concat(values=[img1,img2],axis=)
sess=tf.Session()
#sess.run(tf.initialize_all_variables())
sess.run(tf.global_variables_initializer())
print("out1=",type(img))
#转化为numpy数组
#通过.eval函数可以把tensor转化为numpy类数据
img_numpy=img.eval(session=sess)
print("out2=",type(img_numpy))
#转化为tensor
img_tensor= tf.convert_to_tensor(img_numpy)
print("out2=",type(img_tensor))

mxnet
from mxnet import nd
x = nd.ones((,))
a = x.asnumpy()
(type(a), a)
nd.array(a)

pytorch
import torch
import numpy as np
np_data = np.arange().reshape((, ))
torch_data = torch.from_numpy(np_data)
tensor2array = torch_data.numpy()
print(
    '\nnumpy array:', np_data,          # [[  ], [  ]]
    '\ntorch tensor:', torch_data,      #       \n         [torch.LongTensor of size 2x3]
    '\ntensor to array:', tensor2array, # [[  ], [  ]]
)