In the past five years, Python has become a popular data the scientific community. Therefore, it is slowly taking over R- "statistical terms" - as the preferred tool of many tools. The recently released Stack Overflow Developer Survey 2018 show, Python is an important programming language, its use in the industry will continue to increase. Python rise of shocking, but not surprising. Its versatility, combined with the efficiency and ease of use, so you can more easily build data science. You can also use a wealth of Python libraries to handle all scientific data-related tasks, from basic Web crawl to the complex task of training deep learning model.
In this article, we will introduce some of the areas most popular and widely used Python libraries and its applications.
Web Crawl
With the help of a web browser, web crawling is popular using the HTTP protocol from the network information extraction technology. The two most common Web crawler is based on Python.
1.Beautiful Soup
Beautiful Soup is a popular Python library to extract information from HTML and XML files. It provides a unique and easy way to navigate, search, and modify data analysis, which can save you unnecessary work time. It applies to both versions of Python, that is, 2.7 and 3.x, and very easy to use.
- Tip willing yard
Beautiful Soup URL: https://pypi.org/project/beau...
2.Scrapy
Scrapy written in Python is a free open source framework. Although developed for Web crawling, but it can also be used as a conventional Web crawlers and use a different API to extract data. Follow frameworks like Django "Do not repeat yourself" philosophy, reptiles Scrapy contains a set of self-contained, each and every reptile that follow specific instructions specific targets.
- Tip willing to code
Scrapy URL: https://scrapy.org/
Scientific Computing and Data Analysis
Arguably the most common tasks of scientific data by providing a unique library for the data processing and analysis, and mathematical calculations, thus proving very valuable for data scientists.
3.NumPy
Python is the most popular NumPy scientific computing library, which is part of a larger Python for scientific computing stack, called SciPy (discussed below). In addition to the use of linear algebra and other mathematical functions, and it can be used as a container or a multidimensional array common data having an arbitrary data type.
NumPy seamless integration language (e.g., C / C ++), and because it supports a variety of data types, it is also applicable to various databases.
- Tip willing to code
NumPy URL: http://www.numpy.org/
4.SciPy
SciPy is a Python-based framework that includes mathematical, scientific computing and data analysis of open source libraries. SciPy Library is a collection for advanced math and statistics algorithms and tools. SciPy stack includes the following libraries:
· NumPy - Python package for numerical calculations
· SciPy - one of the core package SciPy stack for signal processing, optimization, and advanced statistics
· matplotlib - for data visualization popular Python libraries
· SymPy - symbolic math and algebra libraries
· pandas - for data manipulation and analysis of Python libraries
· iPython - the code used to run a Python-based interactive console
- Tip willing to code
SciPy URL: https://www.scipy.org/index.html
5.Pandas
pandas Python package is a widely used and effective tool to provide data structures, data manipulation and analysis. It is a widely used tool for quantitative analysis, and found a lot of applications in algorithmic trading and risk analysis.
It has a large community of dedicated users, regularly updated pandas to get new API changes, performance updates and bug fixes.
- Tip willing to code
pandas URL: https://pandas.pydata.org/
Machine learning and deep learning
Python is better than all the other languages in machine learning and deep learning model for achieving efficient, just by virtue of its diverse, effective and easy-to-use set of libraries. In this section, we will see some of the most popular and most commonly used Python library for machine learning and deep learning:
6.Scikit-learn
scikit-learn for data mining, analysis of the most popular Python libraries and machine learning. It uses NumPy, SciPy matplotlib and functional building, and are commercially available. You can use scikit-learn to achieve a variety of machine learning techniques, such as classification, regression, clustering, and very easy to install.
- Tip willing to code
scikit-learn Web site: https://scikit-learn.org/stable/
7.Tensorflow
Tensorflow is a Python-based framework for the use of a plurality of CPU or GPU effective depth of machine learning and learning. Supported by Google, originally developed by the research team of Google Brain, machine intelligence framework is widely used in the world. It got the support of a large number of active users, and is widely used in various industrial fields of advanced machine learning, from manufacturing and retailing to health care and smart cars.
- Tip willing to code
Tensorflow URL: https://www.tensorflow.org/
8.Keras
Keras is a Python-based neural network API, provides a simplified interface makes it easy to train and deploy your depth learning model. It supports a variety of depths learning framework, such as Tensorflow, Deeplearning4j CNTK and very user-friendly and, following the modular approach supports calculate the CPU and GPU. If you want the depth of the learning process easier and more effective, then the library is definitely worth a try!
- Tip willing to code
Keras URL: https://keras.io/
9.PyTorch
PyTorch is one of the newest members of the Python series of deep learning, it is a powerful GPU support neural network modeling library. Although still in beta, but the project was supported by great men such as Facebook and Twitter. PyTorch built on the depth of the library Torch Another popular architecture in order to achieve more efficient tensor calculation and dynamic neural network.
- Tip willing to code
PyTorch URL: https://pytorch.org/
Natural Language Processing
Natural language processing involves the design process, analyze and interpret human language, spoken or written system. Python provides a unique library, to perform various tasks, such as the use of structured and unstructured text, predictive analysis, and so on.
10.NLTK
NLTK is a popular language processing Python library. It provides easy-to-use interface for a variety of NLP tasks, such as text classification, labeling, textual parsing, semantic reasoning and so on. It is an open source, community-driven projects, and support for Python 2 and Python 3.
- Tip willing to code
NLTK URL: http://www.nltk.org/
11.SpaCy
SpaCy is another natural language processing library based on high-level Python and the Cython. It supports a wide range depth learning libraries and frameworks, such as Tensorflow and PyTorch. Use SpaCy, you can relatively easily build complex statistical model NLP. SpaCy easy to install and use, extracted and analyzed in terms of large-scale text information proved to be very useful.
- Tip willing to code
SpaCy URL: https://spacy.io/
data visualization
Data visualization is a widely used in science and technology data, and for transmitting information to visually analyze and valuable business insight through graphs, charts, and reports dashboards. Python provides many popular database for efficient data description. Some of these are listed below:
12.matplotlib
matplotlib is the most popular data visualization Python library that allows enterprise-level 2D and 3D graphics. Use matplotlib, you can use a few lines of code to build a different type of visualization, such as histograms, bar charts, scatter plots, and so on. Matplotlib of popularity comparable with R acclaimed ggplot2, Matplotlib can run seamlessly on all the Python console, including iPython and Jupyter laptops, to provide you with all the necessary tools to create and share data visualization needs.
- Tip willing to code
matplotlib URL: https://matplotlib.org/
13. Seaborn
Seaborn is a Python-based data visualization library, which originated from matplotlib. In addition to providing an attractive and insightful data visualization addition, seaborn also provide strong support to other Python libraries, such as NumPy and pandas.
- Tip willing to code
Seaborn URL: https://seaborn.pydata.org/in...
14. Bokeh
Bokeh is a Python-based interactive data visualization library. It is designed to provide D3.js elegant style graphics and visualization, mainly run on modern Web browser. In addition to the ability to create various visual Outsider, Bokeh also supports real-time data sets of large-scale interaction and visualization.
- Tip willing to code
Bokeh URL: https://bokeh.pydata.org/en/l...
15. Plotly
Plotly Python library is a widely used around the world for the production of publication-quality graphs. Use Plotly, you can easily build interactive dashboards, scatter plots, histograms, candlestick charts, heat maps, and a lot of other data visualization. With excellent interactivity, deployment and publishing capabilities, Plotly can be used in different fields, mainly financial and geospatial industries, for effective data description.
- Tip willing to code
Plotly URL: https://plot.ly/python/
Python provides an extensive set of libraries for each task associated with the scientific data, each library is equipped with a unique feature, you can quickly and easily complete the task. Although there are many Python libraries, but we can choose the 15 libraries according to their popularity, usefulness and the value they bring.
We will certainly encounter many difficulties when learning python, as well as the pursuit of new technologies, here's what we recommend learning Python buckle qun: 784758214, here is the python learner gathering place! ! At the same time, he was a senior development engineer python, python script from basic to web development, reptiles, django, data mining and other projects to combat zero-based data are finishing. Given to every little python partner! Daily share some methods of learning and the need to pay attention to small details
