轉載自:http://www.52nlp.cn/python-網頁爬蟲-文字處理-科學計算-機器學習-資料探勘
曾經因為NLTK的緣故開始學習Python,之後漸漸成為我工作中的第一輔助指令碼語言,雖然開發語言是C/C++,但平時的很多文字資料處理任務都交給了Python。離開騰訊創業後,第一個作品課程圖譜也是選擇了Python系的Flask框架,漸漸的將自己的絕大部分工作交給了Python。這些年來,接觸和使用了很多Python工具包,特別是在文字處理,科學計算,機器學習和資料探勘領域,有很多很多優秀的Python工具包可供使用,所以作為Pythoner,也是相當幸福的。其實如果仔細留意微博,你會發現很多這方面的分享,自己也Google了一下,發現也有同學總結了“Python機器學習庫”,不過總感覺缺少點什麼。最近流行一個詞,全棧工程師(full stack engineer),作為一個苦逼的創業者,天然的要把自己打造成一個full stack engineer,而這個過程中,這些Python工具包給自己提供了足夠的火力,所以想起了這個系列。當然,這也僅僅是拋磚引玉,希望大家能提供更多的線索,來彙總整理一套Python網頁爬蟲,文字處理,科學計算,機器學習和資料探勘的兵器譜。
一、Python網頁爬蟲工具集
一個真實的專案,一定是從獲取資料開始的。無論文字處理,機器學習和資料探勘,都需要資料,除了通過一些渠道購買或者下載的專業資料外,常常需要大家自己動手爬資料,這個時候,爬蟲就顯得格外重要了,幸好,Python提供了一批很不錯的網頁爬蟲工具框架,既能爬取資料,也能獲取和清洗資料,我們也就從這裡開始了:
1. Scrapy
Scrapy, a fast high-level screen scraping and web crawling framework for Python.
鼎鼎大名的Scrapy,相信不少同學都有耳聞,課程圖譜中的很多課程都是依靠Scrapy抓去的,這方面的介紹文章有很多,推薦大牛pluskid早年的一篇文章:《Scrapy 輕鬆定製網路爬蟲》,歷久彌新。
官方主頁:http://scrapy.org/
Github內碼表: https://github.com/scrapy/scrapy
2. Beautiful Soup
You didn’t write that awful page. You’re just trying to get some data out of it. Beautiful Soup is here to help. Since 2004, it’s been saving programmers hours or days of work on quick-turnaround screen scraping projects.
讀書的時候通過《集體智慧程式設計》這本書知道Beautiful Soup的,後來也偶爾會用用,非常棒的一套工具。客觀的說,Beautifu Soup不完全是一套爬蟲工具,需要配合urllib使用,而是一套HTML/XML資料分析,清洗和獲取工具。
官方主頁:http://www.crummy.com/software/BeautifulSoup/
3. Python-Goose
Html Content / Article Extractor, web scrapping lib in Python
Goose最早是用Java寫得,後來用Scala重寫,是一個Scala專案。Python-Goose用Python重寫,依賴了Beautiful Soup。前段時間用過,感覺很不錯,給定一個文章的URL, 獲取文章的標題和內容很方便。
Github主頁:https://github.com/grangier/python-goose
二、Python文字處理工具集
從網頁上獲取文字資料之後,依據任務的不同,就需要進行基本的文字處理了,譬如對於英文來說,需要基本的tokenize,對於中文,則需要常見的中文分詞,進一步的話,無論英文中文,還可以詞性標註,句法分析,關鍵詞提取,文字分類,情感分析等等。這個方面,特別是面向英文領域,有很多優秀的工具包,我們一一道來。
1. NLTK — Natural Language Toolkit
NLTK is a leading platform for building Python programs to work with human language data. It provides easy-to-use interfaces to over 50 corpora and lexical resources such as WordNet, along with a suite of text processing libraries for classification, tokenization, stemming, tagging, parsing, and semantic reasoning, and an active discussion forum.
搞自然語言處理的同學應該沒有人不知道NLTK吧,這裡也就不多說了。不過推薦兩本書籍給剛剛接觸NLTK或者需要詳細瞭解NLTK的同學: 一個是官方的《Natural Language Processing with Python》,以介紹NLTK裡的功能用法為主,同時附帶一些Python知識,同時國內陳濤同學友情翻譯了一箇中文版,這裡可以看到:推薦《用Python進行自然語言處理》中文翻譯-NLTK配套書;另外一本是《Python Text Processing with NLTK 2.0 Cookbook》,這本書要深入一些,會涉及到NLTK的程式碼結構,同時會介紹如何定製自己的語料和模型等,相當不錯。
官方主頁:http://www.nltk.org/
Github內碼表:https://github.com/nltk/nltk
2. Pattern
Pattern is a web mining module for the Python programming language.
It has tools for data mining (Google, Twitter and Wikipedia API, a web crawler, a HTML DOM parser), natural language processing (part-of-speech taggers, n-gram search, sentiment analysis, WordNet), machine learning (vector space model, clustering, SVM), network analysis and canvas visualization.
Pattern由比利時安特衛普大學CLiPS實驗室出品,客觀的說,Pattern不僅僅是一套文字處理工具,它更是一套web資料探勘工具,囊括了資料抓取模組(包括Google, Twitter, 維基百科的API,以及爬蟲和HTML分析器),文字處理模組(詞性標註,情感分析等),機器學習模組(VSM, 聚類,SVM)以及視覺化模組等,可以說,Pattern的這一整套邏輯也是這篇文章的組織邏輯,不過這裡我們暫且把Pattern放到文字處理部分。我個人主要使用的是它的英文處理模組Pattern.en, 有很多很不錯的文字處理功能,包括基礎的tokenize, 詞性標註,句子切分,語法檢查,拼寫糾錯,情感分析,句法分析等,相當不錯。
官方主頁:http://www.clips.ua.ac.be/pattern
3. TextBlob: Simplified Text Processing
TextBlob is a Python (2 and 3) library for processing textual data. It provides a simple API for diving into common natural language processing (NLP) tasks such as part-of-speech tagging, noun phrase extraction, sentiment analysis, classification, translation, and more.
TextBlob是一個很有意思的Python文字處理工具包,它其實是基於上面兩個Python工具包NLKT和Pattern做了封裝(TextBlob stands on the giant shoulders of NLTK and pattern, and plays nicely with both),同時提供了很多文字處理功能的介面,包括詞性標註,名詞短語提取,情感分析,文字分類,拼寫檢查等,甚至包括翻譯和語言檢測,不過這個是基於Google的API的,有呼叫次數限制。TextBlob相對比較年輕,有興趣的同學可以關注。
官方主頁:http://textblob.readthedocs.org/en/dev/
Github內碼表:https://github.com/sloria/textblob
4. MBSP for Python
MBSP is a text analysis system based on the TiMBL and MBT memory based learning applications developed at CLiPS and ILK. It provides tools for Tokenization and Sentence Splitting, Part of Speech Tagging, Chunking, Lemmatization, Relation Finding and Prepositional Phrase Attachment.
MBSP與Pattern同源,同出自比利時安特衛普大學CLiPS實驗室,提供了Word Tokenization, 句子切分,詞性標註,Chunking, Lemmatization,句法分析等基本的文字處理功能,感興趣的同學可以關注。
官方主頁:http://www.clips.ua.ac.be/pages/MBSP
5. Gensim: Topic modeling for humans
Gensim是一個相當專業的主題模型Python工具包,無論是程式碼還是文件,我們曾經用《如何計算兩個文件的相似度》介紹過Gensim的安裝和使用過程,這裡就不多說了。
官方主頁:http://radimrehurek.com/gensim/index.html
github內碼表:https://github.com/piskvorky/gensim
6. langid.py: Stand-alone language identification system
語言檢測是一個很有意思的話題,不過相對比較成熟,這方面的解決方案很多,也有很多不錯的開源工具包,不過對於Python來說,我使用過langid這個工具包,也非常願意推薦它。langid目前支援97種語言的檢測,提供了很多易用的功能,包括可以啟動一個建議的server,通過json呼叫其API,可定製訓練自己的語言檢測模型等,可以說是“麻雀雖小,五臟俱全”。
Github主頁:https://github.com/saffsd/langid.py
7. Jieba: 結巴中文分詞
“結巴”中文分詞:做最好的Python中文分片語件 “Jieba” (Chinese for “to stutter”) Chinese text segmentation: built to be the best Python Chinese word segmentation module.
好了,終於可以說一個國內的Python文字處理工具包了:結巴分詞,其功能包括支援三種分詞模式(精確模式、全模式、搜尋引擎模式),支援繁體分詞,支援自定義詞典等,是目前一個非常不錯的Python中文分詞解決方案。
Github主頁:https://github.com/fxsjy/jieba
8. xTAS
xtas, the eXtensible Text Analysis Suite, a distributed text analysis package based on Celery and Elasticsearch.
感謝微博朋友 @大山坡的春 提供的線索:我們組同事之前釋出了xTAS,也是基於python的text mining工具包,歡迎使用,連結:http://t.cn/RPbEZOW。看起來很不錯的樣子,回頭試用一下。
Github內碼表:https://github.com/NLeSC/xtas
三、Python科學計算工具包
說起科學計算,大家首先想起的是Matlab,集數值計算,視覺化工具及互動於一身,不過可惜是一個商業產品。開源方面除了GNU Octave在嘗試做一個類似Matlab的工具包外,Python的這幾個工具包集合到一起也可以替代Matlab的相應功能:NumPy+SciPy+Matplotlib+iPython。同時,這幾個工具包,特別是NumPy和SciPy,也是很多Python文字處理 & 機器學習 & 資料探勘工具包的基礎,非常重要。最後再推薦一個系列《用Python做科學計算》,將會涉及到NumPy, SciPy, Matplotlib,可以做參考。
1. NumPy
NumPy is the fundamental package for scientific computing with Python. It contains among other things:
1)a powerful N-dimensional array object
2)sophisticated (broadcasting) functions
3)tools for integrating C/C++ and Fortran code
4) useful linear algebra, Fourier transform, and random number capabilitiesBesides its obvious scientific uses, NumPy can also be used as an efficient multi-dimensional container of generic data. Arbitrary data-types can be defined. This allows NumPy to seamlessly and speedily integrate with a wide variety of databases.
NumPy幾乎是一個無法迴避的科學計算工具包,最常用的也許是它的N維陣列物件,其他還包括一些成熟的函式庫,用於整合C/C++和Fortran程式碼的工具包,線性代數、傅立葉變換和隨機數生成函式等。NumPy提供了兩種基本的物件:ndarray(N-dimensional array object)和 ufunc(universal function object)。ndarray是儲存單一資料型別的多維陣列,而ufunc則是能夠對陣列進行處理的函式。
2. SciPy:Scientific Computing Tools for Python
SciPy refers to several related but distinct entities:
1)The SciPy Stack, a collection of open source software for scientific computing in Python, and particularly a specified set of core packages.
2)The community of people who use and develop this stack.
3)Several conferences dedicated to scientific computing in Python – SciPy, EuroSciPy and SciPy.in.
4)The SciPy library, one component of the SciPy stack, providing many numerical routines.
“SciPy是一個開源的Python演算法庫和數學工具包,SciPy包含的模組有最優化、線性代數、積分、插值、特殊函式、快速傅立葉變換、訊號處理和影象處理、常微分方程求解和其他科學與工程中常用的計算。其功能與軟體MATLAB、Scilab和GNU Octave類似。 Numpy和Scipy常常結合著使用,Python大多數機器學習庫都依賴於這兩個模組。”—-引用自“Python機器學習庫”
3. Matplotlib
matplotlib is a python 2D plotting library which produces publication quality figures in a variety of hardcopy formats and interactive environments across platforms. matplotlib can be used in python scripts, the python and ipython shell (ala MATLAB®* or Mathematica®†), web application servers, and six graphical user interface toolkits.
matplotlib 是python最著名的繪相簿,它提供了一整套和matlab相似的命令API,十分適合互動式地進行製圖。而且也可以方便地將它作為繪圖控制元件,嵌入GUI應用程式中。Matplotlib可以配合ipython shell使用,提供不亞於Matlab的繪圖體驗,總之用過了都說好。
4. iPython
IPython provides a rich architecture for interactive computing with:
1)Powerful interactive shells (terminal and Qt-based).
2)A browser-based notebook with support for code, text, mathematical expressions, inline plots and other rich media.
3)Support for interactive data visualization and use of GUI toolkits.
4)Flexible, embeddable interpreters to load into your own projects.
5)Easy to use, high performance tools for parallel computing.
“iPython 是一個Python 的互動式Shell,比預設的Python Shell 好用得多,功能也更強大。 她支援語法高亮、自動完成、程式碼除錯、物件自省,支援 Bash Shell 命令,內建了許多很有用的功能和函式等,非常容易使用。 ” 啟動iPython的時候用這個命令“ipython –pylab”,預設開啟了matploblib的繪圖互動,用起來很方便。
官方主頁:http://ipython.org/
四、Python 機器學習 & 資料探勘 工具包
機器學習和資料探勘這兩個概念不太好區分,這裡就放到一起了。這方面的開源Python工具包有很多,這裡先從熟悉的講起,再補充其他來源的資料,也歡迎大家補充。
1. scikit-learn: Machine Learning in Python
scikit-learn (formerly scikits.learn) is an open source machine learning library for the Python programming language. It features various classification, regression and clustering algorithms including support vector machines, logistic regression, naive Bayes, random forests, gradient boosting, k-means and DBSCAN, and is designed to interoperate with the Python numerical and scientific libraries NumPy and SciPy.
首先推薦大名鼎鼎的scikit-learn,scikit-learn是一個基於NumPy, SciPy, Matplotlib的開源機器學習工具包,主要涵蓋分類,迴歸和聚類演算法,例如SVM, 邏輯迴歸,樸素貝葉斯,隨機森林,k-means等演算法,程式碼和文件都非常不錯,在許多Python專案中都有應用。例如在我們熟悉的NLTK中,分類器方面就有專門針對scikit-learn的介面,可以呼叫scikit-learn的分類演算法以及訓練資料來訓練分類器模型。這裡推薦一個視訊,也是我早期遇到scikit-learn的時候推薦過的:推薦一個Python機器學習工具包Scikit-learn以及相關視訊–Tutorial: scikit-learn – Machine Learning in Python
2. Pandas: Python Data Analysis Library
Pandas is a software library written for the Python programming language for data manipulation and analysis. In particular, it offers data structures and operations for manipulating numerical tables and time series.
第一次接觸Pandas是由於Udacity上的一門資料分析課程“Introduction to Data Science” 的Project需要用Pandas庫,所以學習了一下Pandas。Pandas也是基於NumPy和Matplotlib開發的,主要用於資料分析和資料視覺化,它的資料結構DataFrame和R語言裡的data.frame很像,特別是對於時間序列資料有自己的一套分析機制,非常不錯。這裡推薦一本書《Python for Data Analysis》,作者是Pandas的主力開發,依次介紹了iPython, NumPy, Pandas裡的相關功能,資料視覺化,資料清洗和加工,時間資料處理等,案例包括金融股票資料探勘等,相當不錯。
官方主頁:http://pandas.pydata.org/
=====================================================================
分割線,以上工具包基本上都是自己用過的,以下來源於其他同學的線索,特別是《Python機器學習庫》,《23個python的機器學習包》,做了一點增刪修改,歡迎大家補充
=====================================================================
3. mlpy – Machine Learning Python
mlpy is a Python module for Machine Learning built on top of NumPy/SciPy and the GNU Scientific Libraries.
mlpy provides a wide range of state-of-the-art machine learning methods for supervised and unsupervised problems and it is aimed at finding a reasonable compromise among modularity, maintainability, reproducibility, usability and efficiency. mlpy is multiplatform, it works with Python 2 and 3 and it is Open Source, distributed under the GNU General Public License version 3.
官方主頁:http://mlpy.sourceforge.net/
4. MDP:The Modular toolkit for Data Processing
Modular toolkit for Data Processing (MDP) is a Python data processing framework.
From the user’s perspective, MDP is a collection of supervised and unsupervised learning algorithms and other data processing units that can be combined into data processing sequences and more complex feed-forward network architectures.
From the scientific developer’s perspective, MDP is a modular framework, which can easily be expanded. The implementation of new algorithms is easy and intuitive. The new implemented units are then automatically integrated with the rest of the library.
The base of available algorithms is steadily increasing and includes signal processing methods (Principal Component Analysis, Independent Component Analysis, Slow Feature Analysis), manifold learning methods ([Hessian] Locally Linear Embedding), several classifiers, probabilistic methods (Factor Analysis, RBM), data pre-processing methods, and many others.
“MDP用於資料處理的模組化工具包,一個Python資料處理框架。 從使用者的觀點,MDP是能夠被整合到資料處理序列和更復雜的前饋網路結構的一批監督學習和非監督學習演算法和其他資料處理單元。計算依照速度和記憶體需求而高效的執行。從科學開發者的觀點,MDP是一個模組框架,它能夠被容易地擴充套件。新演算法的實現是容易且直觀的。新實現的單元然後被自動地與程式庫的其餘部件進行整合。MDP在神經科學的理論研究背景下被編寫,但是它已經被設計為在使用可訓練資料處理演算法的任何情況中都是有用的。其站在使用者一邊的簡單性,各種不同的隨時可用的演算法,及應用單元的可重用性,使得它也是一個有用的教學工具。”
官方主頁:http://mdp-toolkit.sourceforge.net/
5. PyBrain
PyBrain is a modular Machine Learning Library for Python. Its goal is to offer flexible, easy-to-use yet still powerful algorithms for Machine Learning Tasks and a variety of predefined environments to test and compare your algorithms.
PyBrain is short for Python-Based Reinforcement Learning, Artificial Intelligence and Neural Network Library. In fact, we came up with the name first and later reverse-engineered this quite descriptive “Backronym”.
“PyBrain(Python-Based Reinforcement Learning, Artificial Intelligence and Neural Network)是Python的一個機器學習模組,它的目標是為機器學習任務提供靈活、易應、強大的機器學習演算法。(這名字很霸氣)
PyBrain正如其名,包括神經網路、強化學習(及二者結合)、無監督學習、進化演算法。因為目前的許多問題需要處理連續態和行為空間,必須使用函式逼近(如神經網路)以應對高維資料。PyBrain以神經網路為核心,所有的訓練方法都以神經網路為一個例項。”
6. PyML – machine learning in Python
PyML is an interactive object oriented framework for machine learning written in Python. PyML focuses on SVMs and other kernel methods. It is supported on Linux and Mac OS X.
“PyML是一個Python機器學習工具包,為各分類和迴歸方法提供靈活的架構。它主要提供特徵選擇、模型選擇、組合分類器、分類評估等功能。”
專案主頁:http://pyml.sourceforge.net/
7. Milk:Machine learning toolkit in Python.
Its focus is on supervised classification with several classifiers available:
SVMs (based on libsvm), k-NN, random forests, decision trees. It also performs
feature selection. These classifiers can be combined in many ways to form
different classification systems.
“Milk是Python的一個機器學習工具箱,其重點是提供監督分類法與幾種有效的分類分析:SVMs(基於libsvm),K-NN,隨機森林經濟和決策樹。它還可以進行特徵選擇。這些分類可以在許多方面相結合,形成不同的分類系統。對於無監督學習,它提供K-means和affinity propagation聚類演算法。”
官方主頁:http://luispedro.org/software/milk
http://luispedro.org/software/milk
8. PyMVPA: MultiVariate Pattern Analysis (MVPA) in Python
PyMVPA is a Python package intended to ease statistical learning analyses of large datasets. It offers an extensible framework with a high-level interface to a broad range of algorithms for classification, regression, feature selection, data import and export. It is designed to integrate well with related software packages, such as scikit-learn, and MDP. While it is not limited to the neuroimaging domain, it is eminently suited for such datasets. PyMVPA is free software and requires nothing but free-software to run.
“PyMVPA(Multivariate Pattern Analysis in Python)是為大資料集提供統計學習分析的Python工具包,它提供了一個靈活可擴充套件的框架。它提供的功能有分類、迴歸、特徵選擇、資料匯入匯出、視覺化等”
9. Pyrallel – Parallel Data Analytics in Python
Experimental project to investigate distributed computation patterns for machine learning and other semi-interactive data analytics tasks.
“Pyrallel(Parallel Data Analytics in Python)基於分散式計算模式的機器學習和半互動式的試驗專案,可在小型叢集上執行”
Github內碼表:http://github.com/pydata/pyrallel
10. Monte – gradient based learning in Python
Monte (python) is a Python framework for building gradient based learning machines, like neural networks, conditional random fields, logistic regression, etc. Monte contains modules (that hold parameters, a cost-function and a gradient-function) and trainers (that can adapt a module’s parameters by minimizing its cost-function on training data).
Modules are usually composed of other modules, which can in turn contain other modules, etc. Gradients of decomposable systems like these can be computed with back-propagation.
“Monte (machine learning in pure Python)是一個純Python機器學習庫。它可以迅速構建神經網路、條件隨機場、邏輯迴歸等模型,使用inline-C優化,極易使用和擴充套件。”
官方主頁:http://montepython.sourceforge.net
11. Theano
Theano is a Python library that allows you to define, optimize, and evaluate mathematical expressions involving multi-dimensional arrays efficiently. Theano features:
1)tight integration with NumPy – Use numpy.ndarray in Theano-compiled functions.
2)transparent use of a GPU – Perform data-intensive calculations up to 140x faster than with CPU.(float32 only)
3)efficient symbolic differentiation – Theano does your derivatives for function with one or many inputs.
4)speed and stability optimizations – Get the right answer for log(1+x) even when x is really tiny.
5)dynamic C code generation – Evaluate expressions faster.
6) extensive unit-testing and self-verification – Detect and diagnose many types of mistake.
Theano has been powering large-scale computationally intensive scientific investigations since 2007. But it is also approachable enough to be used in the classroom (IFT6266 at the University of Montreal).
“Theano 是一個 Python 庫,用來定義、優化和模擬數學表示式計算,用於高效的解決多維陣列的計算問題。Theano的特點:緊密整合Numpy;高效的資料密集型GPU計算;高效的符號微分運算;高速和穩定的優化;動態生成c程式碼;廣泛的單元測試和自我驗證。自2007年以來,Theano已被廣泛應用於科學運算。theano使得構建深度學習模型更加容易,可以快速實現多種模型。PS:Theano,一位希臘美女,Croton最有權勢的Milo的女兒,後來成為了畢達哥拉斯的老婆。”
12. Pylearn2
Pylearn2 is a machine learning library. Most of its functionality is built on top of Theano. This means you can write Pylearn2 plugins (new models, algorithms, etc) using mathematical expressions, and theano will optimize and stabilize those expressions for you, and compile them to a backend of your choice (CPU or GPU).
“Pylearn2建立在theano上,部分依賴scikit-learn上,目前Pylearn2正處於開發中,將可以處理向量、影象、視訊等資料,提供MLP、RBM、SDA等深度學習模型。”