Author: admin yep

Posted on

8 Ways to Stay on Top of the Latest Trends in Data Science

https://static1.makeuseofimages.com/wordpress/wp-content/uploads/2023/06/documents-on-wooden-surface.jpg

Data science is constantly evolving, with new papers and technologies coming out frequently. As such, data scientists may feel overwhelmed when trying to keep up with the latest innovations.

MUO video of the day

SCROLL TO CONTINUE WITH CONTENT

However, with the right tips, you can stay current and remain relevant in this competitive field. Thus, here are eight ways to stay on top of the latest trends in data science.

1. Follow Data Science Blogs and Newsletters

Data science blogs are a great way to brush up on the basics while learning about new ideas and technologies. Several tech conglomerates produce high-quality blog content where you can learn about their latest experiments, research, and projects. Great examples are Google, Facebook, and Netflix blogs, so waste no time checking them out.

Alternatively, you can look into online publications and individual newsletters. Depending on your experience level and advancement in the field, these blogs may address topics you’d find more relatable. For example, Version Control for Jupyter Notebook is easier for a beginner to digest than Google’s Preference learning for cache eviction.

You can find newsletters by doing a simple search, but we’d recommend Data Elixir, Data Science Weekly, and KDnuggets News, as these are some of the best.

2. Listen to Data Science Podcasts and Watch YouTube Videos

Podcasts are easily accessible and a great option when you’re pressed for time and want to get knowledge on the go. Listening to podcasts exposes you to new data science concepts while letting you carry out other activities simultaneously. Also, using interviews with experts in the field, some podcasts offer a window into the industry and let you learn from professionals’ experiences.

On the other hand, YouTube is a better alternative for audio-visual learners and has several videos at your disposal. Channels like Data School and StatQuest with Josh Starmer cover a wide range of topics for both aspiring and experienced data scientists. They also touch on new trends and methods, so following these channels is a good idea to keep current.

It’s easy to get lost in a sea of podcasts and videos, so carefully select detailed videos and the best podcasts for data science. This way, you can acquire accurate knowledge from the best creators and channels.

3. Learn Data Science Skills and Concepts From Courses and Books

Online courses allow learning from data science academics and experts, who condense their years of experience into digestible content. Recent courses cover several data science necessities, from hard-core machine learning to starting a career in data science without a degree. They may not be cheap, but they are well worth their cost in the value they give.

Additionally, books play an important role as well. Reading current data science books can help you learn new techniques, understand real-world data science applications, and develop critical thinking and problem-solving skills. These books explain in-depth data science concepts you may not find elsewhere.

Such books include The Data Science Handbook, Data Science on the Google Cloud Platform, and Think Bayes. You should also check out a few data science courses on sites like Coursera and Udemy.

4. Meet Industry Experts and Enthusiasts From Events and Communities

Attending conferences ushers you into an environment of like-minded individuals you can connect with. Although talking to strangers may feel uncomfortable, you will learn so much from the people at these events. By staying home, you will likely miss out on networking, job opportunities, and modern techniques like deep learning methods.

Furthermore, presentations allow you to observe other projects and familiarize yourself with the latest trends. Seeing what big tech companies are up to is encouraging and educative, and you can always take away something from them to apply in your work.

Data science events can be physical or virtual. Some good data science events to consider are the Open Data Science Conference (ODSC), Data Science Salon, and the Big Data and Analytics Summit.

5. Participate in Data Science Competitions and Hackathons

Data science hackathons unite data scientists to develop models that solve real-world problems within a specified time frame. They can be hosted by various platforms, such as Kaggle, DataHack, or UN Big Data Hackathon.

Participating in hackathons enhances your mastery and accuracy and exposes you to the latest data science tools and popular techniques for building models. Regardless of your results, competing with other data scientists in hackathons offers valuable insights into the latest advancements in data science.

Consider participating in the NERSC Open Hackathon, BNL Open Hackathon, and other virtual hackathons. Also, don’t forget to register for physical hackathons that may be happening near your location.

6. Contribute to Data Science Open Source or Social Good Projects

Contributing to open-source data science projects lets you work with other data scientists in development. From them, you’ll learn new tools and frameworks used by the data science community, and you can study project codes to implement in your work.

Furthermore, you can collaborate with other data scientists with different perspectives in an environment where exchanging ideas, feedback, and insights is encouraged. You can discover the latest techniques data science professionals use, industry standards, best practices, and how they keep up with data science trends.

First, search for repositories tagged with the data science topic on GitHub or Kaggle. Once you discover a project, consider how to contribute, regardless of your skill level, and start collaborating with other data scientists.

Following data science thought leaders and influencers on social media keep you informed about the latest data science trends. This way, you can learn about their views on existing subject matters and up-to-date news on data science trends. Additionally, it allows you to inquire about complicated subjects and get their reply.

You can take it a step further and follow Google, Facebook, Apple, and other big tech companies on Twitter. This gives you the privilege of knowing tech trends to expect, not only limited to data science.

Kirk Borne, Ronald van Loon, and Ian Goodfellow are some of the biggest names in the data science community. Start following them and big tech companies on Twitter and other social media sites to stay updated.

8. Share Your Data Science Work and Insights

Sharing your work lets you get feedback and suggestions from other data scientists with different experience levels and exposure. Their comments, questions, and critiques can help you stay up-to-date with the latest trends in data science.

You can discover trendy ideas, methods, tools, or resources you may not have known before by listening to their suggestions. For example, a person may unknowingly use an outdated version of Python until he posts his work online and someone points it out.

Sites like Kaggle and Discord have several data science groups through which you can share your work and learn. After signing up and joining a group, start asking questions and interacting with other data scientists. Prioritize knowledge, remember to be humble, and try to build mutually beneficial friendships with other data scientists.

Be a Lifelong Learner in Data Science

Continuous learning is necessary to remain valuable as a data scientist, but it can be difficult to keep up all by yourself. Consequently, you’ll need to find a suitable community to help you, and Discord is one of the best platforms to find one. Find a server with people in the same field, and continue your learning with your new team.

MakeUseOf

Posted on

The 6 Best Linux Distros for Network Engineers

https://static1.makeuseofimages.com/wordpress/wp-content/uploads/2023/06/man-works-on-laptop-next-to-networking-equipment.jpg

Linux is commonly preferred among network engineers—so if you’ve thought about installing it for your work, you’re not alone.

MUO video of the day

SCROLL TO CONTINUE WITH CONTENT

If you’re a network engineer, it’s easy to wonder which distributions will have the best features for your work. Here are the six best Linux distributions for network engineering:

1. Fedora

Of all the Linux distributions, one of the most highly regarded among network engineers is Fedora—and there’s a simple reason why.

Fedora is an open-source distribution that serves as a community equivalent to Red Hat Enterprise Linux (RHEL). RHEL itself is commonly chosen as the operating system for enterprise-level systems.

As a result, network engineers who use Fedora enjoy a greater level of familiarity with the RHEL systems they may encounter throughout their careers.

Fedora also offers users an incredible arsenal of open-source tools, built-in support for containerized applications, and consistent access to cutting-edge features and software.

Download: Fedora (free)

2. RHEL

rhel 9 desktop view
Image Credit: RedHat/Wikimedia under CC BY-SA 4.0

As one of the most popular enterprise distributions, RHEL is a great option because it is robust and reinforced. Each version of RHEL has a 10-year lifecycle, meaning that you’ll be able to use your chosen version of RHEL (and enjoy little to no compatibility issues) for years.

By using RHEL, you’ll also become familiar with many of the systems you’re likely to encounter on the job.

Many of the qualities of RHEL that make it attractive as an enterprise solution are just as appealing for independent users.

RHEL comes pre-equipped with the SELinux security module, so you will find it easy to get started with managing access controls and system policies. You’ll also have access to tools like Cacti and Snort through the RPM and YUM package managers.

Download: RHEL (free for developers; $179 annually)

3. CentOS Stream

Much like Fedora, CentOS Stream is a distribution that stays in line with the development of RHEL. It serves as the upstream edition of RHEL, meaning that the content in the latest edition of CentOS Stream is likely to appear in RHEL’s next release.

While CentOS Stream may not offer the same stability as Fedora, its enticing inclusion of cutting-edge software makes it worth considering.

CentOS Stream also has a distinct advantage over downstream editions of RHEL following Red Hat’s decision to close public access to the source code of RHEL: it will continue to stay in line with the latest experimental changes considered for the next release of RHEL.

In the future, CentOS Stream is likely to become the best option for anyone seeking an RHEL-adjacent distribution.

Download: CentOS Stream (free)

4. openSUSE

Another powerful and reliable option for network engineers is openSUSE. openSUSE is impressively stable and offers frequent new releases, making it a good option if you prefer to avoid broken packages while still taking advantage of the latest software releases.

Out of the box, you won’t have any issues configuring basic network settings through YaST (Yet another Setup Tool). Many of the packages that come preinstalled with openSUSE can provide you with incredible utility.

Wicked is a powerful network configuration framework, for example, while Samba is perfect for enabling file-sharing between Linux and Windows systems. You won’t have any trouble installing the right tool for a job with openSUSE’s Zypper package manager.

Download: openSUSE (free)

5. Debian

viewing settings and browser in debian
Image Credit: Граймс/Wikimedia under CC BY-SA 4.0

Debian is a widely-renowned Linux distribution known for being incredibly stable and high-performance. Several branches of Debian are available, including Debian Stable (which is extremely secure and prioritizes stability) and Debian Unstable (which is more likely to break but provides access to the newest cutting-edge releases of software).

One of the biggest advantages of using Debian for network engineering is that it has an incredible package-rich repository with over 59,000 different software packages.

If you’re interested in trying out the newest niche and experimental tools in networking and cybersecurity, an installation of Debian will provide you with total access.

Download: Debian (free)

6. Kali Linux

kali desktop view
Image Credit: Gased Basek/Wikimedia under CC BY-SA 4.0

As a distribution designed for penetration testing, Kali Linux comes with a massive variety of preinstalled tools that network engineers are certain to find useful. Wireshark offers tantalizing information about packets moving across a network, Nmap provides useful clues about network security, and SmokePing provides interesting visualizations of network latency.

Not all of the software packaged with Kali Linux is useful for network engineers, but luckily, new Kali installations are completely customizable. You should plan out what packages you intend to use in advance so that you can avoid installing useless packages and keep your Kali system minimally cluttered.

Download: Kali Linux (free)

Familiarize Yourself With Your New Networking Distro

While some Linux distributions are better suited to network engineers, almost any Linux distribution can be used with the right software and configurations.

You should test out software like Nmap and familiarize yourself with networking on your new Linux distro so that lack of familiarity doesn’t become an obstacle later on.

MakeUseOf

Posted on

Laravel Migrations: “Table already exists” After Foreign Key Failed

https://laraveldaily.com/storage/437/Copy-of-Copy-of-Copy-of-ModelpreventLazyLoading();-(4).png

If you create foreign keys in your migrations, there may be a situation that the table is created successfully, but the foreign key fails. Then your migration is “half successful”, and if you re-run it after the fix, it will say “Table already exists”. What to do?

The Problem: Explained

First, let me explain the problem in detail. Here’s an example.

Schema::create('teams', function (Blueprint $table) {
    $table->id();
    $table->string('name');
    $table->foreignId('team_league_id')->constrained();
    $table->timestamps();
});

The code looks good, right? Now, what if the referenced table “team_leagues” doesn’t exist? Or maybe it’s called differently? Then you will see this error in the Terminal:

2023_06_05_143926_create_teams_table ..................................................................... 20ms FAIL
 
Illuminate\Database\QueryException
 
  SQLSTATE[HY000]: General error: 1824 Failed to open the referenced table 'team_leagues'
  (Connection: mysql, SQL: alter table `teams` add constraint `teams_team_league_id_foreign` foreign key (`team_league_id`) references `team_leagues` (`id`))

But that is only part of the problem. So ok, you realized that the referenced table is called “leagues” and not “team_leagues”. Possible fix options:

  • Either rename the field of “team_league_id” to just “league_id”
  • Or, specify the table ->constrained('leagues')

But the real problem now is the state of the database:

  • The table teams is already created
  • But the foreign key to leagues has failed!

This means there’s no record of this migration success in the “migrations” Laravel system DB table.

Now, the real problem: if you fix the error in the same migration and just run php artisan migrate, it will say, “Table already exists”.

2023_06_05_143926_create_teams_table ...................................................................... 3ms FAIL
 
Illuminate\Database\QueryException
 
  SQLSTATE[42S01]: Base table or view already exists:
  1050 Table 'teams' already exists
  (Connection: mysql, SQL: create table `teams` (...)

So should you create a new migration? Rollback? Let me explain my favorite way of solving this.

Solution: Schema::hasTable() and Separate Foreign Key

You can re-run the migration for already existing tables and ensure they would be created only if they don’t exist with the Schema::hasTable() method.

But then, we need to split the foreignId() into parts because it’s actually a 2-in-1 method: it creates the column (which succeeded) and the foreign key (which failed).

So, we rewrite the migration into this:

if (! Schema::hasTable('teams')) {
    Schema::create('teams', function (Blueprint $table) {
        $table->id();
        $table->string('name');
        $table->unsignedBigInteger('team_league_id');
        $table->timestamps();
    });
}
 
// This may be in the same migration file or in a separate one
Schema::table('teams', function (Blueprint $table) {
    $table->foreign('team_league_id')->constrained('leagues');
});

Now, if you run php artisan migrate, it will execute the complete migration(s) successfully.

Of course, an alternative solution would be to go and manually delete the teams table via SQL client and re-run the migration with the fix, but you don’t always have access to the database if it’s remote. Also, it’s not ideal to perform any manual operations with the database if you use migrations. It may be ok on your local database, but this solution above would be universal for any local/remote databases.

Laravel News Links

Posted on

JK Rowling hater demands she performs a sex act on him, but she ruins his life with a tweet instead

https://www.louderwithcrowder.com/media-library/image.png?id=34222424&width=980

JK Rowling has been the head transphobe in charge for quite some time now. You know how leftist sh*tc*nts get when you express a different opinion on them on the internet. They go right to the -isms and -phobias they accuse you of. In Rowling’s case, she also gets people threatening her life. All for the think-crime of, four years ago, defending someone who was fired for saying boys are boys and girls are girls.

If death threats don’t phase her, what do you think her putting her mouth on your weiner is going to do? One loser found this out the hard way and if he has any friends, I hope they’re all laughing at him now.,

So how did things escalate this quickly? It starts with Maya Forstater, the woman who got fired from her job for "’publishing “offensive” tweets questioning government proposals to allow people to self-identify as the opposite sex.’" Defending this woman is what started Rowling on her road to TERFdom. Forstater had just won a settlement when the organisation "was found to have engaged in unlawful discrimination in its decision not to offer her an employment contract or to renew her visiting fellowship."

Turns out that having a common belief about sex and gender does NOT equal bigotry.

Rowling offered Maya her congratulations.

Which led to Joshua D’silva telling Rowlings to suck his dick. This WAS the link to the tweet. It was literally deleted while I was working on this post after Rowling informed the world that D’silva allegedly had a penis so small, it was barely detectable. How embarrassing.

The whole Rowling thing still cracks me up. She is in no way, shape, or form on our side politically AT ALL. When leftists were "resisting" Trump in 2015-16, they would identify themselves as sects from Hogwarts. We would dare them to read another book.

All it took was one tweet and one single opinion for the left to turn on JK Rowling as if she was literally Voldermort. It’s hilarious. Especially knowing Rowling is sleeping on a giant pile of money as she responds. to each of her haters.

><><><><><><

Brodigan is Grand Poobah of this here website and when he isn’t writing words about things enjoys day drinking, pro-wrestling, and country music. You can find him on the Twitter too.

Facebook doesn’t want you reading this post or any others lately. Their algorithm hides our stories and shenanigans as best it can. The best way to stick it to Zuckerface? Bookmark LouderWithCrowder.com and check us out throughout the day! Also, follow us on Instagram and Twitter.

Louder With Crowder

Posted on

If The Mandalorian Was The A-Team

https://theawesomer.com/photos/2023/06/mandalorian_a_team_t.jpg

If The Mandalorian Was The A-Team

Link

The Mandalorian has some great action. So did The A-Team. So Nebulous Bee thought it would be fun to combine the two. This edit reimagines the credits for the Star Wars series in the style of the 1980s hit series, with Bo-Katan, Paz Vizsla, The Armorer, and Din Djarin standing in for Templeton Peck  B.A. Baracus, Hannibal Smith, and Howling Mad Murdock.

The Awesomer

Posted on

The implications of this technology are staggering

http://img.youtube.com/vi/vSPhhw-2ShI/0.jpg

 

I was astonished to read of the wide-ranging implications of a new laser weeding technology now available to farmers.

Carbon Robotics is now shipping its LaserWeeder to farms around the United States; the machine uses the power of lasers and robotics to rid fields of weeds … The LaserWeeder can eliminate over 200,000 weeds per hour and offer up to 80% cost savings in weed control. 

. . .

The LaserWeeder is a 20-foot-wide unit comprised of three rows of 10 lasers that are pulled behind a tractor.

Thirty lasers are at work as the unit travels across a field destroying weeds "with millimeter accuracy, skipping the plant and killing the weed," said Mikesell. 

The LaserWeeder "does the equivalent work of about 70 people," he continued.

. . .

The technology "makes for a much more consistent growing process and adds a bunch of health to your yield. You get big yield improvements because you’re not damaging the crops with herbicides."

There’s more at the link.

Here’s a publicity video from Carbon Robotics showing the LaserWeeder in action.

The economic implications for farmers and farm workers are mind-boggling.

  • The workers normally hired to manage weeds in crops won’t be needed any more – or, at any rate, far fewer of them.  That’s a huge money-saver for farmers, but how many workers will end up unemployed, with no jobs available to replace those they’ve lost?  What will that do to the unemployment rate overall?
  • I’ve no idea how much per acre farmers normally spend on herbicides, but it’s got to add up.  It probably varies from region to region.  If those expenses are no longer needed, the robotic/laser technology of the LaserWeeder becomes that much more affordable.
  • What will this mean for fertilizers and other input costs?  If crops are no longer threatened by weed incursion, will farmers still need as much fertilizer to obtain high yields, or will the absence of weeds – and the saving of time and money through not having to fight them – mean that less fertilizer can be used, because overall crop productivity will be higher even without it?
  • Can this technology be scaled according to the size of farm and type of crop?  The video above shows a big machine in a big field.  Can a smaller machine be made at a lower cost?  Can smaller farms use it cost-effectively?  Can the technology be adapted to (say) market gardening in greenhouses, rather than fields?  These things may not be possible now, but if they become feasible, they may make even the small-scale, backyard growing of fruit and vegetables much easier and cheaper.  Might we be able to grow a certain proportion of our own food, more practically and affordably than before, thereby reducing our dependence on "Big Ag"?
  • Do these input cost savings mean that farmers (and Big Ag in particular) can/will accept lower prices for their produce, because they’ll have lower input costs to grow them?
  • Can this technology be adapted to (say) gardening in greenhouses and back yards, rather than larger fields?  It may not be possible now, but if it becomes feasible, it may make the small-scale, private cultivation of fruit and vegetables much easier and cheaper.  You might see groups of neighbors hiring or buying such technology to share among themselves, at home or in allotments.
  • Over time, this technology may revolutionize the production of food, thereby addressing some of the "woke" or "green" concerns about modern farming practices.  There’s a lot of concern about the over-use of farm chemicals and resultant pollution problems (see, for example, the so-called "dead zone" in the Gulf of Mexico, caused by such chemicals draining down the Mississippi River and out to sea).  Could such technology help reduce that problem, by needing less fertilizer and/or herbicides?

Just the thought of no longer having to spend hours weeding in the back yard is enormously tempting.  This will bear watching.

Peter

Bayou Renaissance Man

Posted on

This Rig Turns Walls/Windows Into Animated Lite-Brites

https://s3files.core77.com/blog/images/1414093_81_124043_JrOgOXAJJ.jpg

Interior/exterior decoration, 2020s-style: Govee Curtain Lights are essentially a curtain of hanging LED-embedded strips. Measuring 1.5m (5′) wide and 2m (6.6′) tall, this provides a grid of 520 evenly-spaced LEDs that you can program via an app, allowing you to turn walls or windows into gigantic animated Lite-Brites.

You can either choose from stock animations provided by the company, or create your own. The company also boasts of a "Music Mode" that makes the lights move in accordance with music.

If this is your jam, startup Govee paid the YouTuber below to produce this video on setting up and using the product.

All I can think is how dated this is going to look in a few years; I imagine if this style of decoration catches on, the consumer will demand a much higher level of resolution.

Core77