1. Flask
This is a micro-framework written in Python. It does not have validations for forms or a level of database abstraction, but allows you to use third-party libraries for common functions. That’s precisely why it is called a micro-framework. Flask is designed for easy and quick application development, as well as being scalable and lightweight. It is based on the Werkzeug and Jinja2 projects. You can learn more about it in the latest DataFlair article on .
2. Keras
Keras is an open-source neural network library written in Python. It is user-friendly, modular, and extensible, and can work on top of TensorFlow, Theano, PlaidML, or Microsoft Cognitive Toolkit (CNTK). Keras includes everything: templates, target and transfer functions, optimizers, and much more. It also supports convolutional and recurrent neural networks.
Working on the latest open-source project based on Keras — .
This article is translated with support from EDISON Software, which , as well as .
3. SpaCy
This is an open-source software library that specializes in and is written in Python and Cython. While NLTK is more suitable for educational and research purposes, SpaCy's focus is on providing software for production. Additionally, Thinc is SpaCy's machine learning library, which features CNN models for part-of-speech tagging, dependency parsing, and named entity recognition.
4. Sentry
Sentry offers open-source error monitoring hosting so you can detect and sort errors in real-time. Simply install the SDK for your language(s) or framework(s) and get started. It allows you to capture unhandled exceptions, study stack traces, analyze the impact of each issue, track errors across various projects, assign issues, and much more. Using Sentry means fewer errors and more code being shipped.
5. OpenCV
OpenCV is an open-source computer vision and machine learning library. It contains over 2500 optimized algorithms for computer vision tasks such as object detection and recognition, classification of various types of human activities, motion tracking using cameras, creating 3D models of objects, stitching images for high-resolution outputs, and many other tasks. The library is available in multiple languages, including Python, C++, Java, etc.
Stars on GitHub: 39,585
Have you worked on any OpenCV projects? Here’s one —
6. Nilearn
This module is designed for the fast and simple implementation of statistical learning on NeuroImaging data. It enables the use of scikit-learn for multivariate statistics for predictive modeling, classification, decoding, and connectivity analysis. Nilearn is part of the NiPy ecosystem, which is a community dedicated to using Python for neuroimaging data analysis.
Stars on : 549
7. scikit-Learn
Scikit-learn is another open-source Python project. It is a widely recognized machine learning library for Python. Often used with NumPy and SciPy, it offers classification, regression, and clustering — it supports , random forests, gradient boosting, k-means, and DBSCAN. This library is written in Python and Cython.
Stars on GitHub: 37,144
8. PyTorch
PyTorch is yet another open-source machine learning library written in and for Python. It is based on the Torch library and is excellent for areas such as computer vision and natural language processing (NLP). It also has a C++ frontend.
Among many other features, PyTorch offers two high-level functionalities:
- Tensor computations with strong acceleration using GPUs
- Deep neural networks
Stars on GitHub: 31,779
9. Librosa
Librosa is one of the best Python libraries for music and audio analysis. It includes essential components used for extracting information from music. The library is well documented and contains several guides and examples to help with your tasks.
GitHub stars: 3107
Python implementation of an open-source project with Librosa — .
10. Gensim
Gensim is a Python library for topic modeling, document indexing, and similarity search among large corpora. It is aimed at NLP and information retrieval communities. Gensim stands for 'generate similar'. It used to create a shortlist of articles similar to a given article. Gensim is understandable, efficient, and scalable. Gensim provides an effective and simple implementation of unsupervised semantic modeling from plain text.
GitHub stars: 9,870
11. Django
— a high-level Python framework that encourages rapid development and follows the DRY principle (Don't Repeat Yourself). It is a very powerful and the most widely used framework for Python. It is based on the MTV (Model-Template-View) pattern.
GitHub stars: 44,214
12. Face recognition
Face recognition is a popular project on GitHub. It easily recognizes faces and manipulates them using Python / command line and utilizes the simplest facial recognition library in the world. It uses dlib with deep learning for face detection with 99.38% accuracy in the Wild benchmark test.
GitHub stars: 28,267
13. Cookiecutter
Cookiecutter is a command-line utility that can be used to create projects from templates (cookiecutters). One example could be creating a package project from a package project template. These are cross-platform templates, and project templates can be in any language or markup format, such as Python, JavaScript, HTML, Ruby, CoffeeScript, RST, and Markdown. It also allows the use of multiple languages in the same project template.
GitHub stars: 10,291
14. Pandas
Pandas is a data analysis and manipulation library for Python, offering labeled data structures and statistical functions.
GitHub stars: 21,404
An open-source Python project to try Pandas —
15. Pipenv
Pipenv promises to be a production-ready tool aimed at bringing the best of all packaging worlds to the Python ecosystem. Its terminal features beautiful colors and combines Pipfile, pip, and virtualenv into a single command. It automatically creates and manages a virtual environment for your projects and provides users with an easy way to set up their workspace.
Number of stars on Github: 18,322
16. SimpleCoin
This is a Blockchain implementation for cryptocurrency, built in Python, but it is simple, insecure, and incomplete. SimpleCoin is not intended for production use. It is designed for educational purposes only, to make a working blockchain accessible and simplified. It allows you to store mined hashes and exchange them for any supported currency.
Number of stars on Github: 1,343
17. Pyray
This is a 3D rendering library written in vanilla Python. It visualizes 2D, 3D, higher-dimensional objects, and scenes in Python, as well as animations. It finds its place in the realm of created videos, video games, physical simulations, and even beautiful images. Requirements for this include: PIL, numpy, and scipy.
Number of stars on Github: 451
18. MicroPython
MicroPython is Python for microcontrollers. It is an efficient implementation of Python3 that comes with many packages from the standard Python library and is optimized to run on microcontrollers and in constrained environments. The Pyboard is a small electronic board where MicroPython runs on bare metal, allowing it to control various electronic projects.
Stars on : 9,197
19. Kivy
Kivy is a Python library for developing mobile applications and other multitouch applications with a natural user interface (NUI). It features a graphics library, several widget options, an intermediate language Kv for creating custom widgets, support for mouse, keyboard, TUIO, and multitouch input events. This is an open-source library for rapid application development with innovative user interfaces. It is cross-platform, business-friendly, and GPU-accelerated.
Number of stars on Github: 9,930
20. Dash
Dash by Plotly is a web application framework. Built on Flask, Plotly.js, React, and React.js, it allows us to use Python to build dashboards. It enables Python and R models to operate at scale. Dash provides capabilities for creating, testing, deploying, and reporting without the need for DevOps, JavaScript, CSS, or CronJobs. Dash is performant, customizable, lightweight, and easy to manage. It also has open-source code.
Number of stars on GitHub: 9,883
21. Magenta
Magenta is an open-source research project that focuses on machine learning as a tool in the creative process. It allows for the creation of music and art using machine learning. Magenta is a Python library built on TensorFlow, with utilities for handling raw data, training machine models, and generating new content.
22. Mask R-CNN
This is an implementation of Mask R-CNN in Python 3, TensorFlow, and Keras. The model takes each instance of an object on the raster and generates bounding boxes and segmentation masks for it. It uses a Feature Pyramid Network (FPN) and ResNet101 backbone. The code is easy to extend. This project also offers the Matterport3D dataset on reconstructed 3D spaces captured by clients…
Number of stars on GitHub: 14,055
23. TensorFlow Models
This repository contains various models implemented in TensorFlow — both official and research models. It also has samples and tutorials. Official models use high-level TensorFlow APIs. Research models are those built in TensorFlow by researchers to support them or to address inquiries and requests.
Number of stars on GitHub: 57,745
24. Snallygaster
Snallygaster is a way to organize project board issues. With it, you can set up a project dashboard on GitHub, optimizing and automating workflows. It allows for sorting tasks, planning projects, automating workflows, tracking progress, sharing status, and finally, completion. Snallygaster can scan for secret files on HTTP servers — it looks for files accessible on web servers that should not be public and may pose a security threat.
Number of stars on GitHub: 1,477
25. Statsmodels
This , which complements scipy for statistical computations, including descriptive statistics, as well as estimations and inferences for statistical models. It contains classes and functions for this purpose. It also allows us to conduct statistical tests and analyze statistical data.
Number of stars on Github: 4,246
26. WhatWaf
This is an advanced firewall detection tool that we can use to determine if a web application firewall is present. It detects firewalls in web applications and attempts to find one or more bypasses for it on the specified target.
Number of stars on Github: 1,300
27. Chainer
Chainer is , focused on flexibility. It is based on Python and offers differentiated APIs based on the define-by-run approach. Chainer also provides high-level object-oriented APIs for building and training neural networks. It is a powerful, flexible, and intuitive framework for neural networks.
Number of stars on Github: 5,054
28. Rebound
Rebound is a command-line tool. When you receive a compiler error message, it immediately retrieves results from the overflowed stack. To use this, you can execute your file using the rebound command. It is one of the 50 most popular open-source Python projects of 2018. Additionally, it requires Python 3.0 or higher. Supported file types: Python, Node.js, Ruby, Golang, and Java.
Number of stars on Github: 2,913
29. Detectron
Detectron performs state-of-the-art object detection (also implements Mask R-CNN). This software is developed by Facebook AI Research (FAIR), written in Python, and runs on the Caffe2 Deep Learning platform. The goal of Detectron is to provide a high-quality and high-performance codebase for object detection research. It is flexible and implements the following algorithms: Mask R-CNN, RetinaNet, Faster R-CNN, RPN, Fast R-CNN, R-FCN.
Number of stars on Github: 21,873
30. Python-fire
This is a library for automatically generating CLI (command-line interfaces) from any Python object. It also allows you to develop and debug code, explore existing code, or turn someone else's code into a CLI. Python Fire makes it easier to switch between Bash and Python and enhances the use of REPL.
Stars on Github: 15,299
31. Pylearn2
Pylearn2 is a machine learning library built primarily on Theano. Its goal is to facilitate ML research. It allows for the writing of new algorithms and models.
Stars on Github: 2,681
32. Matplotlib
is a 2D plotting library for Python – it generates high-quality publications in various formats.
Stars on Github: 10,072
33. Theano
Theano is a library for manipulating mathematical and matrix expressions. It is also an optimizing compiler. Theano uses -like syntax for expressing computations and compiles them to run on CPU or GPU architectures. It is an open-source Python machine learning library written in Python and CUDA and works on Linux, macOS, and Windows.
Stars on : 8,922
34. Multidiff
Multidiff is designed to facilitate the understanding of machine-oriented data. It helps to visualize differences between a large number of objects by performing diffs on corresponding objects and then displaying them. This visualization allows us to search for patterns in our own protocols or unusual file formats. It is also primarily used for reverse engineering and analyzing binary data.
Stars on Github: 262
35. Som-tsp
This project is dedicated to using self-organizing maps to solve the traveling salesman problem. By using SOM, we find suboptimal solutions to the TSP problem and use the .tsp format for this. TSP is an NP-complete problem, and it becomes increasingly difficult to solve as the number of cities grows.
Stars on Github: 950
36. Photon
Photon is an exceptionally fast web scanner designed for OSINT. It can retrieve URLs, parameterized URLs, intel information, files, secret keys, JavaScript files, regex matches, and subdomains. The extracted information can then be saved and exported in JSON format. Photon is flexible and brilliant. You can also add some plugins to it.
Number of stars on GitHub: 5714
37. Social Mapper
Social Mapper is a social mapping tool that correlates profiles using facial recognition. It does this across various websites on a large scale. Social Mapper automates the search for names and photos on social networks and then attempts to accurately identify and group someone's presence. It then generates a report for human verification. This is useful in the security industry (e.g., for phishing). It supports platforms like LinkedIn, Facebook, Twitter, Google Plus, Instagram, VKontakte, Weibo, and Douban.
Number of stars on GitHub: 2,396
38. Camelot
Camelot is a Python library that helps extract tables from PDF files. It works with text-based PDF files but not with scanned documents. Each table is represented as a pandas DataFrame. Additionally, you can export tables to .json, .xls, .html, or .sqlite.
Number of stars on GitHub: 2415
39. Lector
This is a Qt reader for reading e-books. It supports file formats such as .pdf, .epub, .djvu, .fb2, .mobi, .azw / .azw3 / .azw4, .cbr / .cbz, and .md. Lector features a main window, table view, book view, distraction-free reading mode, annotation support, comic viewing, and settings window. It also supports bookmarks, profile viewing, metadata editor, and an integrated dictionary.
Number of stars on GitHub: 835
40. m00dbot
This is a Telegram bot for self-testing depression and anxiety.
Number of stars on GitHub: 145
41. Manim
This is an animation engine for explaining mathematical videos that can be used to create accurate animations programmatically. It utilizes Python for this purpose.
Number of stars on GitHub: 13,491
42. Douyin-Bot
A bot written in Python for an application similar to Tinder. Developed by creators from China.
Number of stars on GitHub: 5,959
43. XSStrike
This package is for cross-site scripting detection with four custom-written parsers. It also features an intelligent payload generator, a powerful fuzzing engine, and an incredibly fast search module. Instead of just inputting the payload and testing its functionality like other tools, XSStrike identifies responses using multiple parsers and then processes payloads that are guaranteed to work through contextual analysis integrated into the fuzzing engine.
Stars on Github: 7050
44. PythonRobotics
This project is a collection of code for algorithms in Python robotics, as well as algorithms for autonomous navigation.
Stars on Github: 6,746
45. Google Images Download
Google Images Download is a command-line Python program that searches for keywords in Google Images and fetches images for you. It’s a small, dependency-free program if you just need to download up to 100 images for each keyword.
Stars on Github: 5749
46. Trape
Enables real-time tracking and execution of social engineering attacks. It helps reveal how major internet companies can obtain confidential information and monitor users without their knowledge. Trape can also assist in tracking cybercriminals.
Stars on Github: 4256
47. Xonsh
Xonsh is a cross-platform Unix-style command-line language and shell based on Python. It is a superset of Python 3.5+ with additional shell primitives, similar to Bash and IPython. Xonsh runs on Linux, macOS, Windows, and other major systems.
Stars on Github: 3426
48. GIF for CLI
This requires a GIF or a short video, or a query, and using the Tenor GIF API, it converts it into animated ASCII art. It uses ANSI escape sequences for animation and color.
Stars on Github: 2,547
49. Cartoonify
Draw This — a Polaroid camera that can draw cartoons. It uses a neural network for object recognition, the Google Quickdraw dataset, a thermal printer, and a Raspberry Pi. Quick, Draw! is a Google game where players are asked to draw an image of an object/idea, and then the game tries to guess what it represents in less than 20 seconds.
Stars on GitHub: 1760
50. Zulip
Zulip is a real-time group chat application that enhances productivity through multithreaded conversations. Many companies in the Fortune 500 list and open-source projects use it for real-time chat that can handle thousands of messages a day.
Stars on GitHub: 10,432
51. YouTube-dl
This is a command-line program that can download videos from YouTube and some other sites. It is not tied to a specific platform.
Stars on GitHub: 55,868
52. Ansible
This is a simple IT automation system that can handle a variety of tasks: configuration management, application deployment, cloud setup, executing special tasks, network automation, and multi-node orchestration.
Stars on GitHub: 39,443
53. HTTPie
HTTPie is a command-line HTTP client. It simplifies CLI interaction with web services. For the http command, it allows us to send arbitrary HTTP requests with simple syntax and receive colored output. We can use it for testing, debugging, and interacting with HTTP servers.
Stars on GitHub: 43,199
54. Tornado Web Server
This is a web framework, an asynchronous networking library for Python. It uses non-blocking I/O to scale to over a thousand open connections. This makes it a good choice for long requests and WebSockets.
Stars on GitHub: 18,306
55. Requests
Requests is a library that makes sending HTTP/1.1 requests easy. You don’t have to manually add parameters to URLs or encode PUT and POST data.
Stars on GitHub: 40,294
56. Scrapy
Scrapy is a fast, high-level framework for web scraping that you can use to extract structured data from websites. You can also utilize it for data analysis, monitoring, and automated testing.
Number of stars on GitHub: 34,493
Source: habr.com
