1. NumPy
In general, we will begin with a library of science as a list, and NumPy is one of the major repositories in the field. It is designed to handle large multidimensional arrays and matrices, and provides a number of advanced mathematical functions and methods that can be used to perform various operations.
In the past year, the development team has made a number of improvements to the library. In addition to bug fixes and compatibility issues, key changes include style improvements, which are the NumPy object's print format. In addition, some functions can now handle any encoded file, as long as the encoding is supported by Python.
2. SciPy
Another scientific computing core library, SCIPY, is built on NumPy and extends the capabilities of NumPy. The main data structures of scipy are multidimensional arrays, implemented using NumPy. The library provides a number of tools for solving tasks such as linear algebra, probability theory, and integration calculations.
SciPy brings significant improvements through continuous integration with different operating systems, such as new functions and methods, and more importantly, the latest optimizer. In addition, many new Blas and LAPACK functions have been packaged by the development team.
3.Pandas
Pandas is a Python library that provides advanced data structures and a variety of analysis tools. One feature of this library is the ability to convert fairly complex data operations into one or two commands. Pandas provides a number of built-in methods for grouping, filtering, and combining data, as well as providing time-series functionality. All of these methods are executed very fast.
The newly released Pandas library also offers hundreds of new features, enhancements, bug fixes, and API changes. These improvements are related to the ability of pandas to group and sort data, and support custom type operations.
4. Statsmodels
Statsmodels is a Python module that provides many possibilities for statistical data analysis, such as statistical model estimation, operational statistics testing, and so on. You can use it to implement many machine learning methods and explore different drawing possibilities.
The library is evolving, bringing more possibilities. This year brings time series improvements and new counting models Generalizedpoisson, 0 expansion models and Negativebinomialp, as well as new multivariate method factor analysis, Manova and ANOVA repeat measurements.
Visualization of
5. Matplotlib
Matplotlib is a low-level library for creating two-dimensional charts and graphs. You can use it to build a variety of charts, from histograms and scatter plots to non-Cartesian coordinate charts. In addition, many popular drawing libraries have reserved locations for matplotlib and can be used in conjunction with Matplotlib.
The library makes a lot of changes to the drawing style, such as color, size, font, legend, and so on. For example, automatic alignment of the axis legend and a more friendly color ring for patients with color blindness.
6. Seaborn
Seaborn is actually a high-level API built on the Matplotlib library. It provides a more appropriate default option for working with charts. In addition, it provides a rich set of visual galleries, including complex types such as time series, Union diagrams, and violin plots.
The Seaborn update is primarily an issue fix. However, the compatibility between Facetgrid (or Pairgrid) and the enhanced interactive matplotlib backend has improved, adding parameters and options to the visualization.
7. plotly
Plotly is a popular library that can help you easily build complex graphics. The library is designed for interactive Web applications and offers many great visualizations, including contour graphics, ternary graphs, and 3D charts.
This library continues to be enhanced and improved, bringing new graphics and features to support "Multi-Link View", Animation and crosstalk integration.
8. Bokeh
The Bokeh library uses JavaScript widgets to create interactive and scalable visual graphics in the browser. The library provides a variety of graphics, styles, link graph forms of interactivity, adding widgets, defining callbacks, and more useful features.
Bokeh's improved interactive features are commendable, such as rotatable classification tick labels, and small scale tools and custom tooltip field enhancements.
9. Pydot
Pydot is a library for generating complex graphics and non-oriented complex graphics. It is used as an interface for Graphviz, written in Python. We can use it to show the structure of the graph, which is often used when building neural networks and decision tree-based algorithms.
What are the 9 most common data analysis libraries used in Python, and what updates have been made in 2018?