热久久最新网站获取,精品综合久久久久久98,国产一久久香蕉国产线看观看

[zz] Python3中print()函数输出多个变量

杰哥 — Sat, 05 Sep 2020 00:44:00 GMT

�W�一�U�方法：(x��) print("变量1", file_name, "变量2", new_name)

print("变量1", file_name, "变量2", new_name)

�W�二�U�方法：(x��) print("变量1�Q?s 变量2�Q?s" % (file_name, new_name))

print("变量1�Q?s 变量2�Q?s" % (file_name, new_name))

�W�三�U�方法：(x��).format()

print('变量1�Q�{0} 变量2�Q�{1}'.format(file_name, new_name))


print('变量1�Q�{0} 变量2�Q�{1}'.format(file_name, new_name))
# 需要注意的是，python是从0开始计敎ͼ�所以第一个{}里要�?开始，依次1�?…

Reference:

https://blog.csdn.net/zhangvalue/article/details/103841029

杰哥 2020-09-05 08:44 发表评论

[zz] 3 个相见恨晚的 Google Colaboratory 奇技

杰哥 — Wed, 13 Feb 2019 15:56:00 GMT

原文出处�Q?/span>

https://www.kdnuggets.com/2018/02/essential-google-colaboratory-tips-tricks.html

试想一下，如果有个免费的在�U�云端��^収ͼ�既可以不用安�?TensorFlow 直接使用�Q�又可以实现 GPU 加速训�l�，那该是多好的一件事情。你没听错，�q�种好事��实存在�Q�今天我��重��介�l�一个谷歌推出的免费的云端工��P��(x��)Colaboratory�?/span>

Colaboratory 是一�?Google 研究��目�Q�旨在帮助传播机器学�?f��n)培训和研究成果�?/span>它是一�?Jupyter �W�记本环�?/span>�Q�不需要进行�Q何设�|�就可以使用�Q��ƈ且完全在云端�q�行。Colaboratory �W�记本存储在 Google 云端��盘中，�q�且可以�׃�n�Q�就如同�(zh��n)��?Google 文档或表��g��栗��Colaboratory 可免费��用�?/span>

也就是说�Q�Colaboratory 存储�?Google 云端��盘中，我们可以�?Google 云端��盘里直接编�?Jupyter Notebook�Q�在�U��用深度学�?f��n)框�?TensorFlow �q�训�l�我们的��经�|�络了。超炫！

�q�里�?3 个��o(h��)人相见恨晚的奇技淫��y来简化它的��用，分别是：(x��)使用免费�?GPU、安装库、上传�ƈ使用数据文�g�?/span>

下面分别介绍�Q?/span>

1. 使用免费�?GPU

在打开�?Jupyter Notebook 中，选择菜单�?#8220;代码执行�E�序�Q�Runtime�Q?/strong>”�Q?#8220;更改�q�行�c�d��Q�Change runtime type�Q?/strong>”�Q�这时将看到以下弹出�H�口�Q?/span>

��保“��g加速器�Q�Hardware accelerator�Q?#8221;讄��?GPU�Q�默认�ؓ(f��) CPU�Q��?span style="color: #444444; font-family: "Microsoft Yahei", "Helvetica Neue", Helvetica, Arial, sans-serif; background-color: #ffffff;">讄��完毕后点��M��存�?/span>

值得注意的是��认�W�记本处�?strong>已连�?/strong>的状态：(x��)

��(g��)查是否真的开启了 GPU�Q�即当前�q�接��C��GPU实例�Q�，可以直接�?Jupyter Notebook 中运行以下命�?/span>�Q?/span>

import tensorflow as tf
device_name = tf.test.gpu_device_name()
if device_name != '/device:GPU:0':
  raise SystemError('GPU device not found')
print('Found GPU at: {}'.format(device_name))

若输��Z��下语句，则表明已�l��用了 GPU 实例�?/span>

Found GPU at: /device:GPU:0

但是�Q�由于在�U?GPU 资源有限�Q�有时候可能会(x��)出现下面的问题提�C�：(x��)

��p��P��h��允许你一�ơ最多持�l��?12 ��时的免�?GPU�?/span>

2. 安装�?/span>

目前�Q�在 Google Colaboratory 中安装的软�g�q�不是持久的�Q�意味着每次重新�q�接实例旉��需要重新安装。但是，Colab 已经默认安装了需要有用的库，安装新的库也�q��难事�Q�方法也有好几种�?/span>

但需要注意的是，安装��M��需要从源代码构建的软�g可能需要很长的旉��?/span>

Colab 同时支持 pip �?apt 包管理器。无论�?zh��n)�使用的是哪一个，��C��要在命��o(h��)前面加上�W�号 “!”�?/span>

# Install Keras with pip
!pip install -q keras
import keras

>>> Using TensorFlow backend.

# Install GraphViz with apt
!apt-get install graphviz -y

3. 上传�q��用数据文�?/span>

我们一般都需要在 Colab �W�记本中使用数据�Q�对吧？你可以��?wget 之类的工具从�|�络上获取数据，但是如果你有一些本地文�Ӟ��想上传到你的��h��盘中的 Colab 环境里�ƈ使用它们�Q�该怎么做呢�Q?/span>

很简单，只需 3 步即可实玎ͼ�

首先使用以下命��o(h��)调用�W�记本中的文仉��择器：(x��)

from google.colab import files
uploaded = files.upload()

�q�行之后�Q�我们就�?x��)发现单�?cell 下出��C��“选择文�g”按钮�Q?/span>

�q�样��可以直接选择你想上传的文件啦�Q?/span>

选择文�g后，使用以下�q�代�Ҏ(gu��)��上传文�g以查扑օ�键名�Q�命令如下：(x��)

for fn in uploaded.keys():
  print('User uploaded file "{name}" with length {length} bytes'.format(name=fn, length=len(uploaded[fn])))

例如待上传的�?iris.csv 文�g�Q�若�q�行没有问题的话�Q�应该出现类��g��面的提示语句�Q?/span>

User uploaded file "iris.csv" with length 3716 bytes

最后，��׃��用以下命令将文�g的内容加载到 Pandas �?DataFrame 中了�Q?/span>

import pandas as pd
import io
df = pd.read_csv(io.StringIO(uploaded['iris.csv'].decode('utf-8')))
print(df)

�q�种上传文�g的方法是不是很简单呢�Q�当�?d��ng)��上传和��用数据文件还有其它的��?gu��)��Q�但是我发现�q�一�Ҏ(gu��)��最��单明了�?/span>

以上��是关于 Google Calaboratory �?3 个非常实用的技巧，赶紧��试一下吧�Q?/span>

最后，可能有的读者朋友对 Google Calaboratory 不太了解的，可以查看我之前写的一��文章：(x��)如何在免费云端运�?Python 深度学习(f��n)框架�Q?/strong>链接如下�Q?/span>

https://redstonewill.com/1493/

也可以点�?strong>阅读原文查看�Q?/span>

Reference:
https://mp.weixin.qq.com/s/vLnBFPZNvIjs4R7d4_6M_g

杰哥 2019-02-13 23:56 发表评论

杰哥 — Fri, 01 Feb 2019 17:51:00 GMT

分��n一��文章：(x��)The Meaning of Underscores in Python�?/p>
本文介绍了Python中单下划�U�和双下划线�Q?dunder"�Q�的各种含义和命名约定，名称修饰�Q�name mangling�Q�的工作原理�Q�以�?qi��ng)它如何影响你自��q��Python�c�R�?/p>
单下划线和双下划�U�在Python变量和方法名�U�C��都各有其含义。有一些含义仅仅是依照�U�定�Q�被视作是对�E�序员的提示 - 而有一些含义是由Python解释器严格执行的�?/p>
如果你想知道“Python变量和方法名�U�C��单下划线和双下划�U�的含义是什么？”�Q�我�?x��)尽我所能在�q�里��Z��解答�?/p>
在本文中�Q�我��讨��Z��下五�U�下划线模式和命名约定，以及(qi��ng)它们如何影响Python�E�序的行为：(x��)
单前��g��划线�Q�_var
单末��下划线�Q�var_
双前��g��划线�Q�__var
双前导和末尾下划�U�：(x��)__var__
单下划线�Q�_
在文章结��֤��Q�你可以扑ֈ�一个简短的“速查�?#8221;�Q��ȝ��了五�U�不同的下划�U�命名约定及(qi��ng)其含义，以及(qi��ng)一个简短的视频教程�Q�可让你亲��n体验它们的行为�?/p>
让我们马上开始！
1. 单前��g��划线 _var
当涉�?qi��ng)到变量和方法名�U�时�Q�单个下划线前缀有一个约定俗成的含义�?它是对程序员的一个提�C?- 意味着Python�C�֌�一致认为它应该是什么意思，但程序的行�ؓ(f��)不受影响�?/p>
下划�U�前�~�的含义是告知其他�E�序员：(x��)以单个下划线开头的变量或方法仅供内部��用�?该约定在PEP 8中有定义�?/p>
�q�不是Python强制规定的�?Python不像Java那样�?#8220;�U�有”�?#8220;公共”变量之间有很强的区别�?�q�就像有人提��Z��一个小��的下划�U�警告标志，��_(d��)��(x��)
“嘿，�q�不是真的要成�ؓ(f��)�cȝ��公共接口的一部分。不�ȝ��它就好�?#8220;
看看下面的例子：(x��)
class Test: def __init__(self): self.foo = 11 self._bar = 23
如果你实例化此类�Q��ƈ��试讉K��在__init__构造函��C��定义的foo和_bar属性，�?x��)发生什么情况？让我们来看看�Q?/p>
>>> t = Test() >>> t.foo 11 >>> t._bar 23
你会(x��)看到_bar中的单个下划�U��ƈ没有��L��我们“�q�入”�c�dƈ讉K��该变量的倹{�?/p>
�q�是因�ؓ(f��)Python中的单个下划�U�前�~�仅仅是一个约�?- 臛_��相对于变量和�Ҏ(gu��)��名而言�?/p>
但是�Q�前��g��划线的确�?x��)�?ji��ng)响从模块中导入名�U�的方式�?/p>
假设你在一个名为my_module的模块中有以下代码：(x��)
# This is my_module.py: def external_func(): return 23 def _internal_func(): return 42
现在�Q�如果��用通配�W�从模块中导入所有名�U�ͼ�则Python不会(x��)导入带有前导下划�U�的名称�Q�除非模块定义了覆盖此行为的__all__列表�Q�：(x��)
>>> from my_module import * >>> external_func() 23 >>> _internal_func() NameError: "name '_internal_func' is not defined"
��Z��说一下，应该避免通配�W�导入，因�ؓ(f��)它们使名�U�空间中存在哪些名称不清楚�?��Z��清楚赯��Q�坚持常规导入更好�?br />
与通配�W�导入不同，常规导入不受前导单个下划�U�命名约定的影响�Q?/p>
>>> import my_module >>> my_module.external_func() 23 >>> my_module._internal_func() 42
我知道这一点可能有点��o(h��)人困惑�?如果你遵循PEP 8推荐�Q�避免通配�W�导入，那么你真正需要记住的只有�q�个�Q?/p>
单个下划�U�是一个Python命名�U�定�Q�表�C��个名�U�是供内部��用的�?它通常不由Python解释器强制执行，仅仅作�ؓ(f��)一�U�对�E�序员的提示�?/blockquote>
2. 单末��下划线 var_
有时候，一个变量的最合适的名称已经被一个关键字所占用�?因此�Q�像class或def�q�样的名�U�C��能用作Python中的变量名称�?在这�U�情况下�Q�你可以附加一个下划线来解军_��名冲�H�：(x��)
>>> def make_object(name, class): SyntaxError: "invalid syntax" >>> def make_object(name, class_): ... pass
��M��Q�单个末��下划线�Q�后�~��Q�是一个约定，用来避免与Python关键字��生命名冲�H��?PEP 8解释了这个约定�?/p>
3. 双前��g��划线 __var
到目前�ؓ(f��)止，我们所涉及(qi��ng)的所有命名模式的含义�Q�来自于已达成共识的�U�定�?而对于以双下划线开头的Python�cȝ��属性（包括变量和方法）�Q�情况就有点不同了�?/p>
双下划线前缀�?x��)导致Python解释器重写属性名�U�ͼ�以避免子�c�M��的命名冲�H��?/p>
�q�也叫做名称修饰�Q�name mangling�Q?- 解释器更改变量的名称�Q�以便在�c�被扩展的时候不�Ҏ(gu��)��产生冲突�?/p>
我知道这听�v来很抽象�?因此�Q�我�l�合了一个小��的代码�C�Z��来予以说明：(x��)
class Test: def __init__(self): self.foo = 11 self._bar = 23 self.__baz = 23
让我们用内置的dir()函数来看看这个对象的属性：(x��)
>>> t = Test() >>> dir(t) ['_Test__baz', '__class__', '__delattr__', '__dict__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__gt__', '__hash__', '__init__', '__le__', '__lt__', '__module__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', '_bar', 'foo']
以上是这个对象属性的列表�?让我们来看看�q�个列表�Q��ƈ��L��我们的原始变量名�U�foo�Q�_bar和__baz - 我保证你�?x��)注意到一些有��的变化�?br />
self.foo变量在属性列表中昄��为未修改为foo�?/li>
self._bar的行为方式相�?- 它以_bar的�Ş式显�C�在�c�M��?��像我之前说�q�的�Q�在�q�种情况下，前导下划�U�仅仅是一个约定�?�l�程序员一个提�C��已�?/li>
然而，对于self.__baz而言�Q�情�늜��h��有点不同�?当你在该列表中搜索__baz�Ӟ��你会(x��)看不到有�q�个名字的变量�?/li>
__baz��Z��么情况了�Q?/p>
如果你仔�l�观察，你会(x��)看到此对象上有一个名为_Test__baz的属性�?�q�就是Python解释器所做的名称修饰�?它这样做是�ؓ(f��)了防止变量在子类中被重写�?/p>
让我们创建另一个扩展Test�cȝ��c�，�q�尝试重写构造函��C��d��的现有属性：(x��)
class ExtendedTest(Test): def __init__(self): super().__init__() self.foo = 'overridden' self._bar = 'overridden' self.__baz = 'overridden'
现在�Q�你认�ؓ(f��)foo�Q�_bar和__baz的��g��(x��)出现在这个ExtendedTest�cȝ��实例上吗�Q?我们来看一看：(x��)
>>> t2 = ExtendedTest() >>> t2.foo 'overridden' >>> t2._bar 'overridden' >>> t2.__baz AttributeError: "'ExtendedTest' object has no attribute '__baz'"
�{�一下，当我们尝试查看t2 .__ baz的值时�Q��ؓ(f��)什么我们会(x��)得到AttributeError�Q?名称修饰被再�ơ触发了�Q?事实证明�Q�这个对象甚��x��有__baz属性：(x��)
>>> dir(t2) ['_ExtendedTest__baz', '_Test__baz', '__class__', '__delattr__', '__dict__', '__dir__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__gt__', '__hash__', '__init__', '__le__', '__lt__', '__module__', '__ne__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', '__weakref__', '_bar', 'foo', 'get_vars']
正如你可以看到__baz变成_ExtendedTest__baz以防止意外修改：(x��)
>>> t2._ExtendedTest__baz 'overridden'
但原来的_Test__baz�q�在�Q?/p>
>>> t2._Test__baz 42
双下划线名称修饰对程序员是完全透明的�?下面的例子证实了�q�一点：(x��)
class ManglingTest: def __init__(self): self.__mangled = 'hello' def get_mangled(self): return self.__mangled >>> ManglingTest().get_mangled() 'hello' >>> ManglingTest().__mangled AttributeError: "'ManglingTest' object has no attribute '__mangled'"
名称修饰是否也适用于方法名�U�ͼ� 是的�Q�也适用。名�U�C��C��(x��)影响在一个类的上下文中，以两个下划线字符�Q?dunders"�Q�开头的所�?/span>名称�Q?/p>
class MangledMethod: def __method(self): return 42 def call_it(self): return self.__method() >>> MangledMethod().__method() AttributeError: "'MangledMethod' object has no attribute '__method'" >>> MangledMethod().call_it() 42
�q�是另一个也��o(h��)人惊讶的�q�用名称修饰的例子：(x��)
_MangledGlobal__mangled = 23 class MangledGlobal: def test(self): return __mangled >>> MangledGlobal().test() 23
在这个例子中�Q�我声明了一个名为_MangledGlobal__mangled的全局变量。然后我在名为MangledGlobal的类的上下文中访问变量。由于名�U�C��饎ͼ�我能够在�cȝ��test()�Ҏ(gu��)��内，以__mangled来引用_MangledGlobal__mangled全局变量�?br />
Python解释器自动将名称__mangled扩展为_MangledGlobal__mangled�Q�因为它以两个下划线字符开头。这表明名称修饰不是专门与类属性关联的。它适用于在�c�M��下文中��用的两个下划�U�字�W�开头的��M��名称�?br />
有很多要吸收的内容吧�?br />
老实��_(d��)��q�些例子和解释不是从我脑子里�y�出来的。我作了一些研�I�和加工才弄出来。我一直��用Python�Q�有很多�q�了�Q�但是像�q�样的规则和�Ҏ(gu��)��情况�q�不��L��现在脑��里�?br />
有时候程序员最重要的技能是“模式识别”�Q�而且知道在哪里查阅信息。如果�?zh��n)�在这一点上感到有点不知所措，请不要担心。慢慢来�Q�试试这��文章中的一些例子�?br />
让这些概念完全沉��怸�来，以便你能够理解名�U�C��饰的��M��思�\�Q�以�?qi��ng)我向�(zh��n)�展示的一些其他的行�ؓ(f��)。如果有一天你和它们不期而遇�Q�你�?x��)知道在文档中按什么来查�?/p>
4. 双前导和双末��下划线 _var_
也许令�h惊讶的是�Q�如果一个名字同时以双下划线开始和�l�束�Q�则不会(x��)应用名称修饰�?由双下划�U�前�~�和后�~�包围的变量不�?x��)被Python解释器修改：(x��)
class PrefixPostfixTest: def __init__(self): self.__bam__ = 42 >>> PrefixPostfixTest().__bam__ 42
但是�Q�Python保留了有双前导和双末��下划线的名�U�ͼ�用于�Ҏ(gu��)��用途�?�q�样的例子有�Q�__init__对象构造函敎ͼ�或__call__ --- 它��得一个对象可以被调用�?/p>
�q�些dunder�Ҏ(gu��)��通常被称为神奇方�?- 但Python�C�֌�中的许多人（包括我自己）都不喜欢�q�种�Ҏ(gu��)��?/p>
最好避免在自己的程序中使用以双下划�U�（“dunders”�Q�开头和�l�尾的名�U�ͼ�以避免与��来Python语言的变化��生冲�H��?/p>
5.单下划线 _
按照�?f��n)惯�Q�有时候单个独立下划线是用作一个名字，来表�C�某个变量是临时的或无关紧要的�?br />
例如�Q�在下面的��@环中�Q�我们不需要访问正在运行的索引�Q�我们可以��?#8220;_”来表�C�它只是一个��(f��)时��|��(x��)
>>> for _ in range(32): ... print('Hello, World.')
你也可以在拆�?unpacking)表达式中��单个下划线用作“不关心的”变量�Q�以忽略特定的倹{�?同样�Q�这个含义只�?#8220;依照�U�定”�Q��ƈ不会(x��)在Python解释器中触发�Ҏ(gu��)��的行为�?单个下划�U�仅仅是一个有效的变量名称�Q�会(x��)有这个用途而已�?/p>
在下面的代码�C�Z��中，我将汽�R元组拆分为单独的变量�Q�但我只寚w��色和里程值感兴趣�?但是�Q��ؓ(f��)了��拆分表达式成功运行，我需要将包含在元�l�中的所有值分配给变量�?在这�U�情况下�Q?#8220;_”作�ؓ(f��)占位�W�变量可以派上用场：(x��)
>>> car = ('red', 'auto', 12, 3812.4) >>> color, _, _, mileage = car >>> color 'red' >>> mileage 3812.4 >>> _ 12
除了用作临时变量之外�Q?#8220;_”是大多数Python REPL中的一个特�D�变量，它表�C�由解释器评估的最�q�一个表辑ּ�的结果�?/p>
�q�样��很方便了，比如你可以在一个解释器�?x��)话中访问先前计��的�l�果�Q�或者，你是在动态构建多个对象�ƈ与它们交互，无需事先�l�这些对象分配名字：(x��)
>>> 20 + 3 23 >>> _ 23 >>> print(_) 23 >>> list() [] >>> _.append(1) >>> _.append(2) >>> _.append(3) >>> _ [1, 2, 3]
Python下划�U�命名模�?- ��结
以下是一个简短的��结�Q�即“速查�?#8221;�Q�罗列了我在本文中谈到的五种Python下划�U�模式的含义�Q?/p>
你想更深入了解学�?f��n)Python知识体系�Q�你可以看一下我们花费了一个多月整理了上百��时的几百个知识点体�p�d��容：(x��)
【超全整理】《Python自动化全能开发从入门到精通》python基础教程�W�记
发布�?2018-04-26
Python

Reference:
https://zhuanlan.zhihu.com/p/36173202

杰哥 2019-02-02 01:51 发表评论

杰哥 — Thu, 03 Jan 2019 04:01:00 GMT

Python3中map函数
在Python2中map函数�?x��)返回一个list列表�Q�如代码�Q?/div>
>>> def f(x, y): return (x, y)
>>> l1 = [ 0, 1, 2, 3, 4, 5, 6 ]
>>> l2 = [ 'Sun', 'Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat' ]
�q�回�l�果如下�Q?/div>
>>> map(f, l1, l2)
[(0, 'Sun'), (1, 'Mon'), (2, 'Tue'), (3, 'Wed'), (4, 'Thu'), (5, 'Fri'), (6, 'Sat')]
但是�Q�在Python3中返回结果如下：(x��)
>>> map(f1, l1, l2)
如果惌��得到Python2的那�U�结果，卌��回list列表�Q�那么必��ȝ��list作用于map�Q�如下：(x��)
>>> list(map(f1, l1, l2))
[(0, 'Sun'), (1, 'Mon'), (2, 'Tue'), (3, 'Wed'), (4, 'Thu'), (5, 'Fri'), (6, 'Sat')]

reference:
https://blog.csdn.net/u010376788/article/details/48230257

杰哥 2019-01-03 12:01 发表评论

[zz]python求数字位数的�Ҏ(gu��)��

杰哥 — Sat, 29 Dec 2018 06:00:00 GMT

�W�一�U�：(x��)
利用str()函数��数字�{化成字符�Ԍ��再利用len()函数判断位长�?/p>
1 a=Int(raw_input("the number you want type in:")
2 b=len(str(a))
3 print b
�W�二�U�：(x��)
利用�?0取商�Q�通过循环�ơ数判断位数�?/p>
c=0
a=int(raw_input("the number you want type in:"))
while a!=0:
a=a/10
c +=1
print c

Reference:

https://www.cnblogs.com/54Leo/p/6133270.html

杰哥 2018-12-29 14:00 发表评论

colab

杰哥 — Fri, 21 Dec 2018 01:07:00 GMT
You can run the deep learning on the cloud without GPU. Zhengxia says that you can run 01_basics.ipynb using colab. Colab will use your gmail account. Upload your file. Edit -> Notebook settings, Runtime type selects Python 3 and Hardware accelerator selects GPU.

How to create a folder in colab? ��h��搜烦(ch��): colab new folder扑ֈ��q�个链接: https://medium.com/lean-in-women-in-tech-india/google-colab-the-beginners-guide-5ad3b417dfa
Create a Colab Notebook
Open Google Colab.
Click on ‘New Notebook’ and select Python 2 notebook or Python 3 notebook.
OR
Open Google Drive.
Create a new folder for the project.
Click on ‘New’ > ‘More’ > ‘Colaboratory’.
Google Drive -> 我的云端��盘 -> 新徏文�g�? Click on ‘New’ > ‘More’ > ‘Colaboratory’,�q�个怎么操作? 例如选中"02_backprop.ipynb", 打开方式, 兌��更多应用, 选择colab. �q�样在google drive里就可以选中"02_backprop.ipynb", 打开方式, ‘Colaboratory’卛_��.但直接打开Colab, Google drive里面是没�?02_backprop.ipynb", Recent有。暂不折腾这个问�? Zhengxia说colab应该�?x��)�?否则大家都把它当�|�盘使用

02_backprop.ipynb, 最开始Forward propagation上面, device改�ؓ(f��)GPU (device = torch.device('cuda') # Uncomment this to run on GPU),没法�q�行原因? zhengxia says: Edit -> Notepad settings -> select GPU.如果一个新�?sh��)脑�Q�你肯定要自己配�|�下cuda和cudnn�Q�谷歌胦(ch��)大气�_�，相当�q�些都帮你做好了�?/div>

杰哥 2018-12-21 09:07 发表评论

[zz] matplotlib.pyplot 中显�C�图像的两种模式�Q�交互和��d��Q�及(qi��ng)其在Python��d��中的应用

杰哥 — Sun, 16 Dec 2018 22:10:00 GMT

介绍
在��用matplotlib的过�E�中�Q�发��C��能像matlab一样同时开几个�H�口�q�行比较�Q�于是查询得知了交互模式�Q�但是放在脚本里�q�行的适合却��L��一闪而过�Q�图像�ƈ不停留，遂仔�l�阅��d��理解了一下文档，��C��解决办法�Q�问题比较简单，仅供菜鸟参考�?/p>
python可视化库matplotlib有两�U�显�C�模式：(x��)
��d��Q�block�Q�模�?/li>
交互�Q�interactive�Q�模�?/li>
在Python Consol命��o(h��)行中�q�行脚本�Q�默认是��d��模式。而在python IDE中运行脚本，matplotlib默认是交互模式。（使用python命��o(h��)行运行脚本不能同时显�C�Z��同图像）
其中的区别是:
在交互模式下�Q?/p>
plt.plot(x)或plt.imshow(x)是直接出囑փ��Q�不需要plt.show()
如果在脚本中使用ion()命��o(h��)开启了交互模式�Q�没有��用ioff()关闭的话�Q�则囑փ��?x��)一闪而过�Q��ƈ不会(x��)常留。要想防止这�U�情况，需要在plt.show()之前加上ioff()命��o(h��)或者��?span style="box-sizing: border-box; outline: 0px; font-weight: 700; overflow-wrap: break-word;">plt.pause(seconds)廉��昄��?/li>
在阻塞模式下�Q?/p>
打开一个窗口以后必��d��掉才能打开下一个新的窗口。这�U�情况下�Q�默认是不能像Matlab一样同时开很多�H�口�q�行�Ҏ(gu��)��的�?/li>
plt.plot(x)或plt.imshow(x)是直接出囑փ��Q�需要plt.show()后才能显�C�图�?/li>
�C�Z��
下面�q�个例子讲的是如何像matlab一样同时打开多个�H�口昄��囄��或线条进行比较，同时也是在脚本中开启交互模式后囑փ�一闪而过的解军_��法：(x��)
import matplotlib.pyplot as plt plt.ion() # 打开交互模式 # 同时打开两个�H�口昄��囄�� plt.figure() plt.imshow(i1) plt.figure() plt.imshow(i2) # 昄��前关掉交互模�?/span> plt.ioff() plt.show()
Reference:
https://blog.csdn.net/wonengguwozai/article/details/79686062

杰哥 2018-12-17 06:10 发表评论

Error:Cannot start process, the path specified for working directory is not a directory

杰哥 — Mon, 05 Nov 2018 01:59:00 GMT

If I open one file such as "hw3_prob4Sol.py", the code can run. However, there is an error: Cannot start process, the path specified for working directory is not a directory. I can use the following method to solve this problem.
PyCharm change working directory of console
Settings -> Build Execution Deployment -> Console -> Python Console

However, if I change the working directory every time, this gives me much trouble. Zhengxia is inspired by
"I finally solved the problem.
I think it all started because the first project that I opened with pycharm was in my "download" folder, so the working directory was automatically set to a temporal folder by default and although I moved the project to another folder and I manually changed the working directory from the terminal, it was not working.
The solution was creating a new project and giving a correct path to the new project. It seems very easy but it was not that obvious." from https://stackoverflow.com/questions/28202273/working-directory-error/28975990.

He gave me the suggestion: File -> open, open the directory rather than open the file. For example, "GAN\code\generative-models-master", he said that it is better to make the environment of the pycharm projects consistent (pycharm工程默认加蝲一个环境变�?最好同一个工�E�下环境一�?. Then a file ".idea" will be created. In this way, all files of this folder will be seen as a part of this project. In the file ".idea", there will be the historical record, the path setting such as the interpreter. Generally, we do not use this file.
Reference:
https://stackoverflow.com/questions/37818938/pycharm-change-working-directory-of-console

杰哥 2018-11-05 09:59 发表评论

PyCharm skills: indention

杰哥 — Sat, 13 Oct 2018 21:08:00 GMT
Tab: Right indention
Shift + Tab: Left indention
Sui Li says that word, powerpoint and group wiki (foreseer) also use this.

杰哥 2018-10-14 05:08 发表评论

How to read *.mat file in python?

杰哥 — Sat, 13 Oct 2018 20:47:00 GMT
# See P4 of "HW2_545_2018_"
import scipy.io as sio
mnist_49_3000 = sio.loadmat('mnist_49_3000.mat')

杰哥 2018-10-14 04:47 发表评论

How to read *.data in Matlab and Python

杰哥 — Fri, 12 Oct 2018 20:33:00 GMT

See P2 of "HW2_545_2018_"

In Matlab:
z = dlmread('spambase.data',',');
In Python:
import numpy as np
z = np.genfromtxt('spambase.data', dtype=float, delimiter=',')

杰哥 2018-10-13 04:33 发表评论

product in python: np.dot, np.multiply, *

杰哥 — Sat, 06 Oct 2018 00:33:00 GMT

https://docs.scipy.org/doc/numpy-1.15.0/reference/generated/numpy.dot.html
Dot product of two arrays. Specifically,
If both a and b are 1-D arrays, it is inner product of vectors (without complex conjugation).
If both a and b are 2-D arrays, it is matrix multiplication, but using matmul or a @ b is preferred.
If either a or b is 0-D (scalar), it is equivalent to multiply and using numpy.multiply(a, b) or a * b is preferred.
If a is an N-D array and b is a 1-D array, it is a sum product over the last axis of a and b.
If a is an N-D array and b is an M-D array (where M>=2), it is a sum product over the last axis of a and the second-to-last axis of b:
dot(a, b)[i,j,k,m] = sum(a[i,j,:] * b[k,:,m])
What does the last two points mean?
My understanding is that 2-D arrays are a matrix. N-D array is a tensor. The aforementioned example is very clear. Both a and b are 3-D arrays.

杰哥 2018-10-06 08:33 发表评论

Interpreter of Pycharm

杰哥 — Fri, 05 Oct 2018 00:59:00 GMT

Yang Zhou says that there can be infinity python.exe in your computer. Maybe some install numpy while some not.
Some other interpreters on my computer:
C:\Users\think\PycharmProjects\Test\venv\Scripts\python.exe Even if I reinstall my operating system, the files in C:\Users\think\PycharmProjects do not need to be stored since this in an empty project.
E:\OneDrive - ustc.edu\PycharmProjects\Test\venv\Scripts
I have set C:\Users\think\Anaconda3\python.exe as the default interpreter. This interpreter has installed numpy. How can we set this? pycharm -> File -> Default settings
pycharm -> File -> settings -> show all -> existing environment
One interesting fact is that if I name a directory as _numpy and "import numpy as np" will have error. This error can be solved by changing the name of the directory from _numpy to _numpy1.

Reference:
https://blog.csdn.net/chengzihe_lu/article/details/79639804

杰哥 2018-10-05 08:59 发表评论

pycharm 默认布局�׃��

杰哥 — Mon, 27 Aug 2018 21:26:00 GMT
�Ҏ(gu��)��https://python.freelycode.com/article/detail/736/解决�Ҏ(gu��)��: 菜单 Window>restore default layout。但是我的菜单栏没有window。�ؕ点，菜单栏下面一行右键，有Customize Menus and Toolbars�Q?Restore All Defaults。菜单栏��有了Window

杰哥 2018-08-28 05:26 发表评论

matlab errorbar

杰哥 — Thu, 06 Sep 2012 01:28:00 GMT

On the group meeting of Oct 3, 2018, Jiaqi shows his bar. I asked that why his variance is very small. He said that he uses the standard error (https://en.wikipedia.org/wiki/Standard_error), which is standard derivation divided by the square root of the number of trials. According to the notes of my undergraduate course probability:
(1) D(ax+b) = a*a*D(x),
(2) the second line of "中心极限定理". x bar is seen as a random variable,
we have std(x bar) = std(x)/sqrt(n).
I ask Jiaqi which tool he uses to draw his figure. He says he uses python, bar of matplotlib. https://matplotlib.org/api/_as_gen/matplotlib.pyplot.bar.html. The mean of the errors and standard error should be computed first and then be used as the input of xerr and yerr.
I check my TNNLS 2017 paper and I find that I show the std. Maybe next time I can show the standard error.

�q�个函数的意思是�Q?br />ERRORBAR(X,Y,L,U)�Q�X是自变量�Q�Y是因变量�Q�L是Y的变动下限，U是Y的变动上�?/p>
errorbar(X,Y,E)  X是自变量�Q�Y是因变量�Q�E是Y的变动绝对差倹{�?/p>我自己写的例子：(x��)

X=[1 2 3];

Y= [ 0.2 0.4 0.2];

L=[0.1 0.03 0.03];U=[0.1 0.3 0.2];

E= [0.1 0.3 0.2];

figure;

hold on;%一定要有这一句，否则有问�?/div>
bar(X,Y);

errorbar(X,Y,E,'Marker','none','LineStyle','none');

figure;

hold on;%一定要有这一句，否则有问�?/div>
errorbar(X,Y,E);%以下注释的两句用�q�一句就可以�?br />%plot(X,Y);

%errorbar(X,Y,E,'Marker','none','LineStyle','none');

figure;

hold on;%一定要有这一句，否则有问�?/div>
bar(X,Y);

errorbar(X,Y,L,U);%,'Marker','none','LineStyle','none'不能�?/div>
说明�Q�必��M��用hold on�Q�bar是画��q��图， errorbar是花竖线
--------------------------------------------------------------------以下可不�?---------------------------------------------------------------------------
example1�Q?/span>

       x = 1:10;
       y = sin(x);
       e = std(y)*ones(size(x));
       errorbar(x,y,e)

example2�Q?/span>
% 生成�C�Z��数据
x=1:10;
y=cumsum(randn(1,10));
lower = y - (rand(1,10));
upper = y + (rand(1,10));

% �׃��errorbar函数使用相对差值在囑�Ş上绘图，所�?/span>
% 需要将�l�对差��D�{变�ؓ(f��)相对差倹{�?/span>
L = y - lower;
U = upper -y;

% �l�图旉��要设�?hold on
% ��q��?/span>
clf;
figure(1);
hold on;
bar(x,y);
% 此处需要隐藏折�U?/span>
errorbar(x,y,L,U,'Marker','none','LineStyle','none');

% 折线�?/span>
figure(2);
hold('on');
plot( x, y);
errorbar( x, y, L, U);

Reference:
http://blog.sina.com.cn/s/blog_61010ebe0100l9c9.html
http://bio-spring.info/wp/?p=85

杰哥 2012-09-06 09:28 发表评论

Python path; matlab function load, save

杰哥 — Wed, 20 Jun 2012 13:07:00 GMT

matlab函数l(f��)oad 目录�Q�save
目录都用反斜杠，�q�样windows和Linux目录都认�Q�Linux只认反斜�?nbsp;
当前目录�Q?/span>
for s=1:20
filename = num2str(s);
load (filename);
end
当前目录的下一层目�?/span>�Q?/span>
for s=1:20
  %filename = ['./50Train/',num2str(s),'.mat']; %�q�种方式应该更快�Q�省得字�W�串操作
  %filename = ['50Train/',num2str(s)]; %最前面�?/不加也可�?/span>
%filename = ['./50Train/',num2str(s)];
filename = strcat('./50Train/',num2str(s));
load (filename);
end
当前目录的上一层目录：(x��)
for s=1:20
  %filename = ['../50Train/',num2str(s),'.mat'];
%filename = ['../50Train/',num2str(s)];
filename = strcat('../50Train/',num2str(s));
load (filename);
end

dataName = '../Parmprint.mat';%'../Parmprint/'�?/span>Parmprint.mat/'都不�?/span>
例：(x��)caffe-master-20151202\caffe-master\matlab\demo\classification_demo.m, im = imread('../../examples/images/cat.jpg');
----------------------------------------------------------------------
想编一个��@环，每个文�g名依�ơ是1�Q?�Q�等到num
for i =1:num
savefile = [num2str(i),'.mat'];
save(savefile, 'trainIdx', 'testIdx');%trainIdx 和testIdx 可以是其他�Q何变�?br />end

save的名字中含有��数�?/span>
ratio = 0.01;
savefile = ['NFL_ASLAN',num2str(ratio),'.mat'];
save(savefile, 'accuracy');
见我的matlab教材P20

Python path
If I put Homework2Q2.py in another content(目录) and I use the "E:\OneDrive - ustc.edu\Lecture\ustc\Artificial intelligence related\Machine learning\Umich\..."in line 4, there is an error as follows:
SyntaxError: (unicode error) 'unicodeescape' codec can't decode bytes in position 30-31: truncated \uXXXX escape
Zhou Yang says that in Python '\n' has a specific meaning. Similarly, python will think '\u' and '\U' has a specific meaning. Write '\\u' and '\\U' instead. Zhengxia says that there are two other ways to solve this problem:
(1) r'E:\OneDrive - ustc.edu\Lecture\ustc\Artificial intelligence related\Machine learning\Umich\...'
(2) use / rather than \

杰哥 2012-06-20 21:07 发表评论

热久久最新网站获取,精品综合久久久久久98,国产一久久香蕉国产线看观看

[zz] Python3中print()函数输出多个变量

�W�一�U�方法：(x��) print("变量1", file_name, "变量2", new_name)

�W�二�U�方法：(x��) print("变量1�Q?s 变量2�Q?s" % (file_name, new_name))

�W�三�U�方法：(x��).format()

[zz] 3 个相见恨晚的 Google Colaboratory 奇技

1. 单前��g��划线 _var

2. 单末���下划线 var_

3. 双前��g��划线 __var

4. 双前导和双末���下划线 _var_

5.单下划线 _

Python下划�U�命名模�?- ���结

[zz]python求数字位数的�Ҏ(gu��)��

colab

[zz] matplotlib.pyplot 中显�C�图像的两种模式�Q�交互和��d���Q�及(qi��ng)其在Python��d��中的应用

介绍

其中的区别是:

�C�Z��

Error:Cannot start process, the path specified for working directory is not a directory

PyCharm skills: indention

How to read *.mat file in python?

How to read *.data in Matlab and Python

product in python: np.dot, np.multiply, *

Interpreter of Pycharm

pycharm 默认布局�׃��

matlab errorbar

Python path; matlab function load, save

2. 单末��下划线 var_

4. 双前导和双末��下划线 _var_

Python下划�U�命名模�?- ��结

[zz] matplotlib.pyplot 中显�C�图像的两种模式�Q�交互和��d��Q�及(qi��ng)其在Python��d��中的应用