Deep learning-based generative art is being researched. As usual, self-learning is better. Some models, like OpenAI's DALL-E 2, require registration and can only be used online, but others can be used locally, which is usually more enjoyable for curious users. I'll demonstrate the Stable Diffusion model's operation on a standard PC.

Let’s get started.

What It Does

Stable Diffusion uses numerous components:

• A generative model trained to produce images is called a diffusion model. The model is incrementally improving the starting data, which is only random noise. The model has an image, and while it is being trained, the reversed process is being used to add noise to the image. Being able to reverse this procedure and create images from noise is where the true magic is (more details and samples can be found in the paper).

• An internal compressed representation of a latent diffusion model, which may be altered to produce the desired images, is used (more details can be found in the paper). The capacity to fine-tune the generation process is essential because producing pictures at random is not very attractive (as we can see, for instance, in Generative Adversarial Networks).

• A neural network model called CLIP (Contrastive Language-Image Pre-training) is used to translate natural language prompts into vector representations. This model, which was trained on 400,000,000 image-text pairs, enables the transformation of a text prompt into a latent space for the diffusion model in the scenario of stable diffusion (more details in that paper).

This figure shows all data flow:

The weights file size for Stable Diffusion model v1 is 4 GB and v2 is 5 GB, making the model quite huge. The v1 model was trained on 256x256 and 512x512 LAION-5B pictures on a 4,000 GPU cluster using over 150.000 NVIDIA A100 GPU hours. The open-source pre-trained model is helpful for us. And we will.

Install

Before utilizing the Python sources for Stable Diffusion v1 on GitHub, we must install Miniconda (assuming Git and Python are already installed):

wget https://repo.anaconda.com/miniconda/Miniconda3-py39_4.12.0-Linux-x86_64.sh
chmod +x Miniconda3-py39_4.12.0-Linux-x86_64.sh
./Miniconda3-py39_4.12.0-Linux-x86_64.sh
conda update -n base -c defaults conda

Install the source and prepare the environment:

git clone https://github.com/CompVis/stable-diffusion
cd stable-diffusion
conda env create -f environment.yaml
conda activate ldm
pip3 install transformers --upgrade

Download the pre-trained model weights next. HiggingFace has the newest checkpoint sd-v14.ckpt (a download is free but registration is required). Put the file in the project folder and have fun:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Almost. The installation is complete for happy users of current GPUs with 12 GB or more VRAM. RuntimeError: CUDA out of memory will occur otherwise. Two solutions exist.

Running the optimized version

Try optimizing first. After cloning the repository and enabling the environment (as previously), we can run the command:

python3 optimizedSD/optimized_txt2img.py --prompt "hello world" --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Stable Diffusion worked on my visual card with 8 GB RAM (alas, I did not behave well enough to get NVIDIA A100 for Christmas, so 8 GB GPU is the maximum I have;).

Running Stable Diffusion without GPU

If the GPU does not have enough RAM or is not CUDA-compatible, running the code on a CPU will be 20x slower but better than nothing. This unauthorized CPU-only branch from GitHub is easiest to obtain. We may easily edit the source code to use the latest version. It's strange that a pull request for that was made six months ago and still hasn't been approved, as the changes are simple. Readers can finish in 5 minutes:

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available at line 20 of ldm/models/diffusion/ddim.py ().

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available in line 20 of ldm/models/diffusion/plms.py ().

• Replace device=cuda in lines 38, 55, 83, and 142 of ldm/modules/encoders/modules.py with device=cuda if torch.cuda.is available(), otherwise cpu.

• Replace model.cuda() in scripts/txt2img.py line 28 and scripts/img2img.py line 43 with if torch.cuda.is available(): model.cuda ().

Run the script again.

Testing

Test the model. Text-to-image is the first choice. Test the command line example again:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

The slow generation takes 10 seconds on a GPU and 10 minutes on a CPU. Final image:

Hello world is dull and abstract. Try a brush-wielding hamster. Why? Because we can, and it's not as insane as Napoleon's cat. Another image:

Generating an image from a text prompt and another image is interesting. I made this picture in two minutes using the image editor (sorry, drawing wasn't my strong suit):

I can create an image from this drawing:

python3 scripts/img2img.py --prompt "A bird is sitting on a tree branch" --ckpt sd-v1-4.ckpt --init-img bird.png --strength 0.8

It was far better than my initial drawing:

I hope readers understand and experiment.

Stable Diffusion UI

Developers love the command line, but regular users may struggle. Stable Diffusion UI projects simplify image generation and installation. Simple usage:

• Unpack the ZIP after downloading it from https://github.com/cmdr2/stable-diffusion-ui/releases. Linux and Windows are compatible with Stable Diffusion UI (sorry for Mac users, but those machines are not well-suitable for heavy machine learning tasks anyway;).

• Start the script.

Done. The web browser UI makes configuring various Stable Diffusion features (upscaling, filtering, etc.) easy:

V2.1 of Stable Diffusion

I noticed the notification about releasing version 2.1 while writing this essay, and it was intriguing to test it. First, compare version 2 to version 1:

• alternative text encoding. The Contrastive LanguageImage Pre-training (CLIP) deep learning model, which was trained on a significant number of text-image pairs, is used in Stable Diffusion 1. The open-source CLIP implementation used in Stable Diffusion 2 is called OpenCLIP. It is difficult to determine whether there have been any technical advancements or if legal concerns were the main focus. However, because the training datasets for the two text encoders were different, the output results from V1 and V2 will differ for the identical text prompts.

• a new depth model that may be used to the output of image-to-image generation.

• a revolutionary upscaling technique that can quadruple the resolution of an image.

• Generally higher resolution Stable Diffusion 2 has the ability to produce both 512x512 and 768x768 pictures.

The Hugging Face website offers a free online demo of Stable Diffusion 2.1 for code testing. The process is the same as for version 1.4. Download a fresh version and activate the environment:

conda deactivate  
conda env remove -n ldm  # Use this if version 1 was previously installed
git clone https://github.com/Stability-AI/stablediffusion
cd stablediffusion
conda env create -f environment.yaml
conda activate ldm

Hugging Face offers a new weights ckpt file.

The Out of memory error prevented me from running this version on my 8 GB GPU. Version 2.1 fails on CPUs with the slow conv2d cpu not implemented for Half error (according to this GitHub issue, the CPU support for this algorithm and data type will not be added). The model can be modified from half to full precision (float16 instead of float32), however it doesn't make sense since v1 runs up to 10 minutes on the CPU and v2.1 should be much slower. The online demo results are visible. The same hamster painting with a brush prompt yielded this result:

It looks different from v1, but it functions and has a higher resolution.

The superresolution.py script can run the 4x Stable Diffusion upscaler locally (the x4-upscaler-ema.ckpt weights file should be in the same folder):

python3 scripts/gradio/superresolution.py configs/stable-diffusion/x4-upscaling.yaml x4-upscaler-ema.ckpt

This code allows the web browser UI to select the image to upscale:

The copy-paste strategy may explain why the upscaler needs a text prompt (and the Hugging Face code snippet does not have any text input as well). I got a GPU out of memory error again, although CUDA can be disabled like v1. However, processing an image for more than two hours is unlikely:

Stable Diffusion Limitations

When we use the model, it's fun to see what it can and can't do. Generative models produce abstract visuals but not photorealistic ones. This fundamentally limits The generative neural network was trained on text and image pairs, but humans have a lot of background knowledge about the world. The neural network model knows nothing. If someone asks me to draw a Chinese text, I can draw something that looks like Chinese but is actually gibberish because I never learnt it. Generative AI does too! Humans can learn new languages, but the Stable Diffusion AI model includes only language and image decoder brain components. For instance, the Stable Diffusion model will pull NO WAR banner-bearers like this:

V1:

V2.1:

The shot shows text, although the model never learned to read or write. The model's string tokenizer automatically converts letters to lowercase before generating the image, so typing NO WAR banner or no war banner is the same.

I can also ask the model to draw a gorgeous woman:

V1:

V2.1:

The first image is gorgeous but physically incorrect. A second one is better, although it has an Uncanny valley feel. BTW, v2 has a lifehack to add a negative prompt and define what we don't want on the image. Readers might try adding horrible anatomy to the gorgeous woman request.

If we ask for a cartoon attractive woman, the results are nice, but accuracy doesn't matter:

V1:

V2.1:

Another example: I ordered a model to sketch a mouse, which looks beautiful but has too many legs, ears, and fingers:

V1:

V2.1: improved but not perfect.

V1 produces a fun cartoon flying mouse if I want something more abstract:

I tried multiple times with V2.1 but only received this:

The image is OK, but the first version is closer to the request.

Stable Diffusion struggles to draw letters, fingers, etc. However, abstract images yield interesting outcomes. A rural landscape with a modern metropolis in the background turned out well:

V1:

V2.1:

Generative models help make paintings too (at least, abstract ones). I searched Google Image Search for modern art painting to see works by real artists, and this was the first image:

I typed "abstract oil painting of people dancing" and got this:

V1:

V2.1:

It's a different style, but I don't think the AI-generated graphics are worse than the human-drawn ones.

The AI model cannot think like humans. It thinks nothing. A stable diffusion model is a billion-parameter matrix trained on millions of text-image pairs. I input "robot is creating a picture with a pen" to create an image for this post. Humans understand requests immediately. I tried Stable Diffusion multiple times and got this:

This great artwork has a pen, robot, and sketch, however it was not asked. Maybe it was because the tokenizer deleted is and a words from a statement, but I tried other requests such robot painting picture with pen without success. It's harder to prompt a model than a person.

I hope Stable Diffusion's general effects are evident. Despite its limitations, it can produce beautiful photographs in some settings. Readers who want to use Stable Diffusion results should be warned. Source code examination demonstrates that Stable Diffusion images feature a concealed watermark (text StableDiffusionV1 and SDV2) encoded using the invisible-watermark Python package. It's not a secret, because the official Stable Diffusion repository's test watermark.py file contains a decoding snippet. The put watermark line in the txt2img.py source code can be removed if desired. I didn't discover this watermark on photographs made by the online Hugging Face demo. Maybe I did something incorrectly (but maybe they are just not using the txt2img script on their backend at all).

Conclusion

The Stable Diffusion model was fascinating. As I mentioned before, trying something yourself is always better than taking someone else's word, so I encourage readers to do the same (including this article as well;).

Is Generative AI a game-changer? My humble experience tells me:

• I think that place has a lot of potential. For designers and artists, generative AI can be a truly useful and innovative tool. Unfortunately, it can also pose a threat to some of them since if users can enter a text field to obtain a picture or a website logo in a matter of clicks, why would they pay more to a different party? Is it possible right now? unquestionably not yet. Images still have a very poor quality and are erroneous in minute details. And after viewing the image of the stunning woman above, models and fashion photographers may also unwind because it is highly unlikely that AI will replace them in the upcoming years.

• Today, generative AI is still in its infancy. Even 768x768 images are considered to be of a high resolution when using neural networks, which are computationally highly expensive. There isn't an AI model that can generate high-resolution photographs natively without upscaling or other methods, at least not as of the time this article was written, but it will happen eventually.

• It is still a challenge to accurately represent knowledge in neural networks (information like how many legs a cat has or the year Napoleon was born). Consequently, AI models struggle to create photorealistic photos, at least where little details are important (on the other side, when I searched Google for modern art paintings, the results are often even worse;).

• When compared to the carefully chosen images from official web pages or YouTube reviews, the average output quality of a Stable Diffusion generation process is actually less attractive because to its high degree of randomness. When using the same technique on their own, consumers will theoretically only view those images as 1% of the results.

Anyway, it's exciting to witness this area's advancement, especially because the project is open source. Google's Imagen and DALL-E 2 can also produce remarkable findings. It will be interesting to see how they progress.

Dark Web activity? Is it as horrible as they say?

We peruse our phones for hours. Internet has improved our worldview.

However, the world's harshest realities remain buried on the internet and unattainable by everyone.

Browsers cannot access the Dark Web. Browse it with high-security authentication and exclusive access. There are compelling reasons to avoid the dark web at all costs.

1. The Dark Web and I

Darius wrote My Dark Web Story on reddit two years ago. The user claimed to have shared his dark web experience. DaRealEddyYT wanted to surf the dark web after hearing several stories.

He curiously downloaded Tor Browser, which provides anonymity and security.

In the Dark Room, bound

As Darius logged in, a text popped up: “Want a surprise? Click on this link.”

The link opened to a room with a chair. Only one light source illuminated the room. The chair held a female tied.

As the screen read "Let the game begin," a man entered the room and was paid in bitcoins to torment the girl.

The man dragged and tortured the woman.

A danger to safety

Leaving so soon, Darius, disgusted Darius tried to leave the stream. The anonymous user then sent Darius his personal information, including his address, which frightened him because he didn't know Tor was insecure.

After deleting the app, his phone camera was compromised.

He also stated that he left his residence and returned to find it unlocked and a letter saying, Thought we wouldn't find you? Reddit never updated the story.

The story may have been a fake, but a much scarier true story about the dark side of the internet exists.

2. The Silk Road Market

The dark web is restricted for a reason. The dark web has everything illicit imaginable. It's awful central.

The dark web has everything, from organ sales to drug trafficking to money laundering to human trafficking. Illegal drugs, pirated software, credit card, bank, and personal information can be found in seconds.

The dark web has reserved websites like Google. The Silk Road Website, which operated from 2011 to 2013, was a leading digital black market.

The FBI grew obsessed with site founder and processor Ross William Ulbricht.

The site became a criminal organization as money laundering and black enterprises increased. Bitcoin was utilized for credit card payment.

The FBI was close to arresting the site's administrator. Ross was detained after the agency closed Silk Road in 2013.

Two years later, in 2015, he was convicted and sentenced to two consecutive life terms and forty years. He appealed in 2016 but was denied, thus he is currently serving time.

The hefty sentence was for more than running a black marketing site. He was also convicted of murder-for-hire, earning about $730,000 in a short time.

3. Person-buying auctions

Bidding on individuals is another weird internet activity. After a Milan photo shoot, 20-year-old British model Chloe Ayling was kidnapped.

An ad agency in Milan made a bogus offer to shoot with the mother of a two-year-old boy. Four men gave her anesthetic and put her in a duffel bag when she arrived.

She was held captive for several days, and her images and $300,000 price were posted on the dark web. Black Death Trafficking Group kidnapped her to sell her for sex.

She was told two black death foot warriors abducted her. The captors released her when they found she was a mother because mothers were less desirable to sex slave buyers.

In July 2018, Lukasz Pawel Herba was arrested and sentenced to 16 years and nine months in prison. Being a young mother saved Chloe from creepy bidding.

However, it exceeds expectations of how many more would be in such danger daily without their knowledge.

4. Organ sales

Many are unaware of dark web organ sales. Patients who cannot acquire organs often turn to dark web brokers.

Brokers handle all transactions between donors and customers.

Bitcoins are used for dark web transactions, and the Tor server permits personal data on the web.

The WHO reports approximately 10,000 unlawful organ transplants annually. The black web sells kidneys, hearts, even eyes.

To protect our lives and privacy, we should manage our curiosity and never look up dangerous stuff.

While it's fascinating and appealing to know what's going on in the world we don't know about, it's best to prioritize our well-being because one never knows how bad it might get.

Sources

Reddit.com

The Daily Beast

PYMNTS

Commons.erau.edu

The Sun

Investopedia

Startup Talky

Let's see this once-dominant IT corporation crumble.

What's the capital of Kazakhstan? If you don't know the answer, you can probably find it by Googling. Google Search returned results for Nur-Sultan in 0.66 seconds.

Google is the best search engine I've ever used. Did you know another search engine ruled the Internet? I'm sure you guessed Yahoo!

Google's friendly UI and wide selection of services make it my top choice. Let's explore Yahoo's decline.

Yahoo!

YAHOO stands for Yet Another Hierarchically Organized Oracle. Jerry Yang and David Filo established Yahoo.

Yahoo is primarily a search engine and email provider. It offers News and an advertising platform. It was a popular website in 1995 that let people search the Internet directly. Yahoo began offering free email in 1997 by acquiring RocketMail.

According to a study, Yahoo used Google Search Engine technology until 2000 and then developed its own in 2004.

Yahoo! rejected buying Google for $1 billion

Larry Page and Sergey Brin, Google's founders, approached Yahoo in 1998 to sell Google for $1 billion so they could focus on their studies. Yahoo denied the offer, thinking it was overvalued at the time.

Yahoo realized its error and offered Google $3 billion in 2002, but Google demanded $5 billion since it was more valuable. Yahoo thought $5 billion was overpriced for the existing market.

In 2022, Google is worth $1.56 Trillion.

What happened to Yahoo!

Yahoo refused to buy Google, and Google's valuation rose, making a purchase unfeasible.

Yahoo started losing users when Google launched Gmail. Google's UI was far cleaner than Yahoo's.

Yahoo offered $1 billion to buy Facebook in July 2006, but Zuckerberg and the board sought $1.1 billion. Yahoo rejected, and Facebook's valuation rose, making it difficult to buy.

Yahoo was losing users daily while Google and Facebook gained many. Google and Facebook's popularity soared. Yahoo lost value daily.

Microsoft offered $45 billion to buy Yahoo in February 2008, but Yahoo declined. Microsoft increased its bid to $47 billion after Yahoo said it was too low, but Yahoo rejected it. Then Microsoft rejected Yahoo’s 10% bid increase in May 2008.

In 2015, Verizon bought Yahoo for $4.5 billion, and Apollo Global Management bought 90% of Yahoo's shares for $5 billion in May 2021. Verizon kept 10%.

Yahoo's opportunity to acquire Google and Facebook could have been a turning moment. It declined Microsoft's $45 billion deal in 2008 and was sold to Verizon for $4.5 billion in 2015. Poor decisions and lack of vision caused its downfall. Yahoo's aim wasn't obvious and it didn't stick to a single domain.

Hence, a corporation needs a clear vision and a leader who can see its future.

Liked this article? Join my tech and programming newsletter here.

Move people with tiny words.

Apple fanboy here.

• Macs are awesome.

• Their iPhones rock.

• $19 cloths are great.

• $999 stands are amazing.

I love Apple's microcopy even more.

It's like the marketing goddess bit into the Apple logo and blessed the world with microcopy.

I took on a 30-day micro-stalking mission.

Every time I caught myself wasting time on YouTube, I had to visit Apple’s website to learn the secrets of the marketing goddess herself.

We've learned. Golden apples are calling.

Cut the friction

Benefit-first, not commitment-first.

Brands lose customers through friction.

Most brands don't think like customers.

• Brands want sales.

• Brands want newsletter signups.

Here's their microcopy:

• “Buy it now.”

• “Sign up for our newsletter.”

Both are difficult. They ask for big commitments.

People are simple creatures. Want pleasure without commitment.

Apple nails this.

So, instead of highlighting the commitment, they highlight the benefit of the commitment.

Saving on the latest iPhone sounds easier than buying it. Everyone saves, but not everyone buys.

A subtle change in framing reduces friction.

Apple eliminates customer objections to reduce friction.

Less customer friction means simpler processes.

Apple's copy expertly reassures customers about shipping fees and not being home. Apple assures customers that returning faulty products is easy.

Apple knows that talking to a real person is the best way to reduce friction and improve their copy.

Always rhyme

Learn about fine rhyme.

Poets make things beautiful with rhyme.

Copywriters use rhyme to stand out.

Apple’s copywriters have mastered the art of corporate rhyme.

Two techniques are used.

1. Perfect rhyme

Here, rhymes are identical.

2. Imperfect rhyme

Here, rhyming sounds vary.

Apple prioritizes meaning over rhyme.

Apple never forces rhymes that don't fit.

It fits so well that the copy seems accidental.

Add alliteration

Alliteration always entertains.

Alliteration repeats initial sounds in nearby words.

Apple's copy uses alliteration like no other brand I've seen to create a rhyming effect or make the text more fun to read.

For example, in the sentence "Sam saw seven swans swimming," the initial "s" sound is repeated five times. This creates a pleasing rhythm.

Microcopy overuse is like pouring ketchup on a Michelin-star meal.

Alliteration creates a memorable phrase in copywriting. It's subtler than rhyme, and most people wouldn't notice; it simply resonates.

I love how Apple uses alliteration and contrast between "wonders" and "ease".

Assonance, or repeating vowels, isn't Apple's thing.

You ≠ Hero, Customer = Hero

Your brand shouldn't be the hero.

Because they'll be using your product or service, your customer should be the hero of your copywriting. With your help, they should feel like they can achieve their goals.

I love how Apple emphasizes what you can do with the machine in this microcopy.

It's divine how they position their tools as sidekicks to help below.

This one takes the cake:

Dialogue-style writing

Conversational copy engages.

Excellent copy Like sharing gum with a friend.

This helps build audience trust.

Apple does this by using natural connecting words like "so" and phrases like "But that's not all."

Snowclone-proof

The mother of all microcopy techniques.

A snowclone uses an existing phrase or sentence to create a new one. The new phrase or sentence uses the same structure but different words.

It’s usually a well know saying like:

To be or not to be.

This becomes a formula:

To _ or not to _.

Copywriters fill in the blanks with cause-related words. Example:

To click or not to click.

Apple turns "survival of the fittest" into "arrival of the fittest."

It's unexpected and surprises the reader.

So this was fun.

But my fun has just begun.

Microcopy is 21st-century poetry.

I came as an Apple fanboy.

I leave as an Apple fanatic.

Now I’m off to find an apple tree.

Cause you know how it goes.

(Apples, trees, etc.)

This post is a summary. Original post available here.