Wen: idle
Source: Python Technology [official ID: PYTHonAll]

GitHub is one of the most popular sites for programmers to visit every day. It is home to many great open source projects. Today, parson is going to take a look at the Top10 most popular Python projects on GitHub over the past year.

Data acquisition

If you’ve been paying attention to the GitHub documentation, you’ll know that most of the data on GitHub is officially provided with an interface. For example, the data that we’re going to get today can be obtained from this interface.

https://api.github.com/search/repositories?q=language:python+created:%3E2019-12-31&sort=stars&order=desc&per_page=10
Copy the code

As shown above, we only get open source projects in Python that were created later than 2019-12-31, that is, new open source projects created in 2020. Next, we order by stars in reverse order, and take the top ten records to get the data we need.

Because the open source project information returned by this interface is too large, we only take the project name, URL, fork number, star number and watch number.

# Retrieve data
def get_data() :
    base_url = 'https://api.github.com/search/repositories?q=language:python+created:%3E2019-12-31&sort=stars&order=desc&per_page=10'
    response = requests.get(base_url)
    result = response.json()
    data = {}
    for item in result['items']:
        data[item['name']] = [item['html_url'], item['stargazers_count'], item['watchers_count'], item['forks']]
    return data
Copy the code

visualization

# visualization
def show_img() :
    data = get_data()
    names = list(data.keys())
    values = [data[name][1] for name in names]

    bar = (
        Bar()
            .add_xaxis(names[::-1])
            .add_yaxis("Star number", values[::-1])
            .reversal_axis()
            .set_series_opts(label_opts=opts.LabelOpts(position="right"))
            .set_global_opts(
            yaxis_opts=opts.AxisOpts(name_rotate=0, name="Project", axislabel_opts={'interval': -10."rotate": 0}),
            title_opts=opts.TitleOpts(title="2020 GitHub Python TOP 10"))
    )
    bar.render_notebook()
Copy the code

A bar chart is generated for the obtained data according to the number of Stars from large to small, as shown below:

00 Depix

With the popularity of mobile Internet, our life is more and more convenient, food, clothing, housing and transportation are all done by a mobile phone, but mobile phones bring us convenience, but also threaten our personal privacy security. For example, many apps often ask you for real-name authentication, even face recognition and so on.

So much so that now we have become extremely cautious, sometimes send a circle of friends to Mosaic, lest a little privacy leakage.

But how would you feel if I told you that it’s no longer safe to blur out, that the information you want to hide is like running naked.

One of the most popular projects on GitHub is Depix, which claims to erase mosaics and recreate original images. So far, the number of star marks of the project has exceeded 14K.

Above is an official example of the picture, we can see that after using Depix recovery, basically can see most of the content, too horrible.

01 diagrams

As programmers, we probably don’t draw as much as we do. We just document. However, drawing is one of the essential skills of architects. As the saying goes, programmers who do not want to be architects are not qualified programmers. Drawing is something we need to learn.

When it comes to drawing, you must think of various online and offline tools, etc. Have you ever tried drawing in code?

Yes, you heard me right. It’s perfectly possible to draw in code. Using diagrams library, we can draw such as Ali Cloud, AWS, K8S system architecture diagram by the way of code.

The chart is the code, and the code is the chart.

02 EasyOCR

Optical Character Recognition (OCR) is to recognize the characters in the image.

EasyOCR is a library to identify characters in images, and it is full of languages (currently covers 70+ languages, including Chinese, Japanese, Korean, Thai).

03 avatarify

As one of the unique skills of Sichuan opera, face-changing has won many people’s love.

And Avatarify allows you to use video conferencing apps like ZOOM and Skype to put celebrities’ faces on your own. Imagine a video conference with your colleagues. Wouldn’t it be cool to use a celebrity’s face?

04 PaddleOCR

The same OCR library has more than 8.1k + stars, but the PaddleOCR is overshadowed by the EasyOCR.

05 eat_tensorflow2_in_30_days

As a machine learning skill, TensorFlow doesn’t have a low barrier to entry, and while TensorFlow2.0 claims to have made significant improvements to improve the user experience, it’s still not easy to use.

Eat_tensorflow2_in_30_days is an easy way to get started with TensorFlow2.0. As an easier way to get started than the official documentation, it has the following advantages.

This tutorial is designed according to the difficulty of the content, the reader’s retrieval habits, and the hierarchy of TensorFlow itself.

Unlike the lengthy sample code in official documents, this tutorial is designed to be as simple and structured as possible in terms of sample code design, and to make the samples more readable and versatile. Most of the code snippets are ready to use in practice.

It can be seen that the author is very attentively, completely stand in a small white identity Angle to think and write, very suitable for small white follow the rhythm step by step towards the peak.

06 GHunt

OSINT is an OSINT tool that sniffs Google accounts, including but not limited to names, Google ids, YouTube channels, and other Google services.

Officials say GHunt allows security teams to browse data created by Google accounts and even analyze a target’s Google trajectory based solely on email. “White hats and penetration testers” can use GHunt to test whether the emails they find are legitimate.

But all I can think about is that our information is about to leak out again, and, uh, why we have to work so hard to protect our privacy.

07 jd_seckill

This is a Python script created on 2020-12-11 to grab Moutai on JD.com.

Nowadays, stealing moutai and selling Moutai has developed into a complete business chain, because there are huge interests in it. Maotai factory official guide price is 1499 yuan, note, this is the manufacturer to sell the dealer price, ordinary consumers are completely unable to buy.

After the sale, the selling price is around 2,599 yuan. Moreover, moutai, as the no.1 liquor in China, does not worry about selling, so it can be said that moutai distributors are making money.

Now, thanks to the development of the Internet, major e-commerce platforms have launched 1499 limited edition moutai activities in order to retain users, which is basically all seconds to kill. Because after grab can easily sell with 2000+ price, a lot of people want, 500+ ocean to hand.

Jd_seckill is a script used to automatically grab Moutai, but due to force majeure, it has been deleted now, it is magic.

08 yolov5

This is an object detection artifact, in other words, to find the bounding box of the object in the picture, and determine the category of the object in the box. For example, identify the car, cat and so on in the picture. We all know that computer vision is very hot right now, and this is a typical application of computer vision.

09 Bringing-Old-Photos-Back-to-Life

This is an artificial intelligence application initiated by Microsoft to restore old photos. It can bring new life to damaged and old images, including scratch repair, overall color restoration and facial repair, and has earned 7.2k + star so far.

conclusion

Here are ten of the most popular Python projects on GitHub in 2020. Which ones have you used?

Welcome to the wechat official account: Python Technology, here we have personally written 100 days of practical training, a variety of interesting programming practices, a variety of learning materials, and a large group of lovely friends to discuss with each other.