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.

It's cheaper, faster charging, longer lasting, safer, and better for the environment.

Batteries are the future. Next-gen and planet-saving technology, including solar power and EVs, require batteries. As these smart technologies become more popular, we find that our batteries can't keep up. Lithium-ion batteries are expensive, slow to charge, big, fast to decay, flammable, and not environmentally friendly. MIT just created a new battery that eliminates all of these problems.  So, is this the battery of the future? Or is there a catch?

When I say entirely new, I mean it. This battery employs no currently available materials. Its electrodes are constructed of aluminium and pure sulfur instead of lithium-complicated ion's metals and graphite. Its electrolyte is formed of molten chloro-aluminate salts, not an organic solution with lithium salts like lithium-ion batteries.

How does this change in materials help?

Aluminum, sulfur, and chloro-aluminate salts are abundant, easy to acquire, and cheap. This battery might be six times cheaper than a lithium-ion battery and use less hazardous mining. The world and our wallets will benefit.

But don’t go thinking this means it lacks performance.

This battery charged in under a minute in tests. At 25 degrees Celsius, the battery will charge 25 times slower than at 110 degrees Celsius. This is because the salt, which has a very low melting point, is in an ideal state at 110 degrees and can carry a charge incredibly quickly. Unlike lithium-ion, this battery self-heats when charging and discharging, therefore no external heating is needed.

Anyone who's seen a lithium-ion battery burst might be surprised. Unlike lithium-ion batteries, none of the components in this new battery can catch fire. Thus, high-temperature charging and discharging speeds pose no concern.

These batteries are long-lasting. Lithium-ion batteries don't last long, as any iPhone owner can attest. During charging, metal forms a dendrite on the electrode. This metal spike will keep growing until it reaches the other end of the battery, short-circuiting it. This is why phone batteries only last a few years and why electric car range decreases over time. This new battery's molten salt slows deposition, extending its life. This helps the environment and our wallets.

These batteries are also energy dense. Some lithium-ion batteries have 270 Wh/kg energy density (volume and mass). Aluminum-sulfur batteries could have 1392 Wh/kg, according to calculations. They'd be 5x more energy dense. Tesla's Model 3 battery would weigh 96 kg instead of 480 kg if this battery were used. This would improve the car's efficiency and handling.

These calculations were for batteries without molten salt electrolyte. Because they don't reflect the exact battery chemistry, they aren't a surefire prediction.

This battery seems great. It will take years, maybe decades, before it reaches the market and makes a difference. Right?

Nope. The project's scientists founded Avanti to develop and market this technology.

So we'll soon be driving cheap, durable, eco-friendly, lightweight, and ultra-safe EVs? Nope.

This battery must be kept hot to keep the salt molten; otherwise, it won't work and will expand and contract, causing damage. This issue could be solved by packs that can rapidly pre-heat, but that project is far off.

Rapid and constant charge-discharge cycles make these batteries ideal for solar farms, homes, and EV charging stations. The battery is constantly being charged or discharged, allowing it to self-heat and maintain an ideal temperature.

These batteries aren't as sexy as those making EVs faster, more efficient, and cheaper. Grid batteries are crucial to our net-zero transition because they allow us to use more low-carbon energy. As we move away from fossil fuels, we'll need millions of these batteries, so the fact that they're cheap, safe, long-lasting, and environmentally friendly will be huge. Who knows, maybe EVs will use this technology one day. MIT has created another world-changing technology.

A TECHNICAL THOUGHT

The IT giant is seeding the digital Great Renaissance.

Recently, technology has been dull.

We're still fascinated by processing speeds. Wearables are no longer an engineer's dream.

Apple has been quiet and avoided huge announcements. Slowness speaks something. Everything in the spaceship HQ seems to be turning slowly, unlike competitors around buzzwords.

Is this a sign of the impending storm?

Metas stock has fallen while Google milks dumb people. Microsoft steals money from corporations and annexes platforms like Linkedin.

Just surface bubbles?

Is Apple, one of the technology continents, pushing against all others to create a paradigm shift?

The fundamental human right to privacy

Apple's unusual remarks emphasize privacy. They incorporate it into their business models and judgments.

Apple believes privacy is a human right. There are no compromises.

This makes it hard for other participants to gain Apple's ecosystem's efficiencies.

Other players without hardware platforms lose.

Apple delivers new kidneys without rejection, unlike other software vendors. Nothing compromises your privacy.

Corporate citizenship will become more popular.

Apples have full coffers. They've started using that flow to better communities, which is great.

Apple's $2.5B home investment is one example. Google and Facebook are building or proposing to build workforce housing.

Apple's funding helps marginalized populations in more than 25 California counties, not just Apple employees.

Is this a trend, and does Apple keep giving back? Hope so.

I'm not cynical enough to suspect these investments have malicious motives.

The last frontier is the environment.

Climate change is a battle-to-win.

Long-term winners will be companies that protect the environment, turning climate change dystopia into sustainable growth.

Apple has been quietly changing its supply chain to be carbon-neutral by 2030.

“Apple is dedicated to protecting the planet we all share with solutions that are supporting the communities where we work.” Lisa Jackson, Apple’s vice president of environment.

Apple's $4.7 billion Green Bond investment will produce 1.2 gigawatts of green energy for the corporation and US communities. Apple invests $2.2 billion in Europe's green energy. In the Philippines, Thailand, Nigeria, Vietnam, Colombia, Israel, and South Africa, solar installations are helping communities obtain sustainable energy.

Apple is already carbon neutral today for its global corporate operations, and this new commitment means that by 2030, every Apple device sold will have net zero climate impact. -Apple.

Apple invests in green energy and forests to reduce its paper footprint in China and the US. Apple and the Conservation Fund are safeguarding 36,000 acres of US working forest, according to GreenBiz.

Apple's packaging paper is recycled or from sustainably managed forests.

What matters is the scale.

$1 billion is a rounding error for Apple.

These small investments originate from a tree with deep, spreading roots.

Apple's genes are anchored in building the finest products possible to improve consumers' lives.

I felt it when I switched to my iPhone while waiting for a train and had to pack my Macbook. iOS 16 dictation makes writing more enjoyable. Small change boosts productivity. Smooth transition from laptop to small screen and dictation.

Apples' tiny, well-planned steps have great growth potential for all consumers in everything they do.

There is clearly disruption, but it doesn't have to be violent

Digital channels, methods, and technologies have globalized human consciousness. One person's responsibility affects many.

Apple gives us tools to be privately connected. These technologies foster creativity, innovation, fulfillment, and safety.

Apple has invented a mountain of technologies, services, and channels to assist us adapt to the good future or combat evil forces who cynically aim to control us and ruin the environment and communities. Apple has quietly disrupted sectors for decades.

Google, Microsoft, and Meta, among others, should ride this wave. It's a tsunami, but it doesn't have to be devastating if we care, share, and cooperate with political decision-makers and community leaders worldwide.

A fresh Renaissance

Renaissance geniuses Michelangelo and Da Vinci. Different but seeing something no one else could yet see. Both were talented in many areas and could discover art in science and science in art.

These geniuses exemplified a period that changed humanity for the better. They created, used, and applied new, valuable things. It lives on.

Apple is a digital genius orchard. Wozniak and Jobs offered us fertile ground for the digital renaissance. We'll build on their legacy.

We may put our seeds there and see them bloom despite corporate greed and political ignorance.

I think the coming tsunami will illuminate our planet like the Renaissance.

Will Tesla be able to maintain its lead in the EV market for very long?

This is one of the most pressing issues in the American auto sector today. One positive aspect of Tesla is the company's devoted customer base and recognizable name recognition (similar to Apple). It also invests more in research and development per vehicle than its rivals and has a head start in EV production.

Conversely, established automakers like Volkswagen are actively plotting their strategy to surpass Tesla. As the current market leaders, they have decades of experience in the auto industry and are spending billions to catch up.

We've visualized data from the EPA's 2022 Automotive Trends Report to bring you up to speed on this developing story.

Info for the Model Year of 2021

The full production data used in this infographic is for the 2021 model year, but it comes from a report for 2022.

Combined EV and PHEV output is shown in the table below (plug-in hybrid electric vehicle).

It is important to note that Toyota and Stellantis, the two largest legacy automakers in this dataset, only produced PHEVs. Toyota's first electric vehicle, the bZ4X, won't hit the market until 2023.

Stellantis seems to be falling even further behind, despite having enormous unrealized potential in its Jeep and Ram brands. Stellantis CEO Carlos Tavares said in a recent interview that the firm has budgeted $36 billion for electrification and software.

Legacy Brands with the Most Momentum

In the race to develop electric vehicles, some long-standing manufacturers have gotten the jump on their rivals.

Volkswagen, one of these storied manufacturers, has made a significant investment in electric vehicles (EVs) in the wake of the Dieselgate scandal. The company plans to roll out multiple EV models, including the ID.3 hatchback, ID.4 SUV, and ID. Buzz, with the goal of producing 22 million EVs by 2028. (an electric revival of the classic Microbus).

Even Ford is keeping up, having just announced an EV investment of $22 billion between 2021 and 2025. In November of 2022, the company manufactured their 150,000th Mustang Mach-E, and by the end of 2023, they hoped to have 270,000 of them in circulation.

Additionally, over 200,000 F-150 Lightnings have been reserved since Ford announced the truck. The Lightning is scheduled to have a production run of 15,000 in 2022, 55,000 in 2023, and 80,000 in 2024. Ford's main competitor in the electric pickup truck segment, Rivian, is on track to sell 25,000 vehicles by 2022.

Writing for readers:

14 years of daily writing.

I post practically everything on social media. I authored hundreds of articles, thousands of tweets, and numerous volumes to almost no one.

Tens of thousands of readers regularly praise me.

I despised writing. I'm stuck now.

I've learned what readers like and what doesn't.

Here are some essential guidelines for writing with impact:

Readers won't understand your work if you can't.

Though obvious, this slipped me up. Share your truths.

Stories engage human brains.

Showing the journey of a person from worm to butterfly inspires the human spirit.

Overthinking hinders powerful writing.

The best ideas come from inner understanding in between thoughts.

Avoid writing to find it. Write.

Writing a masterpiece isn't motivating.

Write for five minutes to simplify. Step-by-step, entertaining, easy steps.

Good writing requires a willingness to make mistakes.

So write loads of garbage that you can edit into a good piece.

Courageous writing.

A courageous story will move readers. Personal experience is best.

Go where few dare.

Templates, outlines, and boundaries help.

Limitations enhance writing.

Excellent writing is straightforward and readable, removing all the unnecessary fat.

Use five words instead of nine.

Use ordinary words instead of uncommon ones.

Readers desire relatability.

Too much perfection will turn it off.

Write to solve an issue if you can't think of anything to write.

Instead, read to inspire. Best authors read.

Every tweet, thread, and novel must have a central idea.

What's its point?

This can make writing confusing.

️ Don't direct your reader.

Readers quit reading. Demonstrate, describe, and relate.

Even if no one responds, have fun. If you hate writing it, the reader will too.