Disclaimer : This is not financial advice for any investment.

TL;DR

• You might not have realized it, but as we move toward net zero carbon emissions, the globe is already at war.

• According to the Paris Agreement of COP26, 64% of nations have already declared net zero, and the issue of carbon reduction has already become so important for businesses that it affects their ability to survive. Furthermore, the time when carbon emission standards will be defined and controlled on an individual basis is becoming closer.

• Since 2017, the market for carbon credits has experienced extraordinary expansion as a result of widespread talks about carbon credits. The carbon credit market is predicted to expand much more once net zero is implemented and carbon emission rules inevitably tighten.

Hello! Ben here from Nonce Classic. Nonce Classic has recently confirmed the tremendous growth potential of the carbon credit market in the midst of a major trend towards the global goal of net zero (carbon emissions caused by humans — carbon reduction by humans = 0 ). Moreover, we too believed that the questions and issues the carbon credit market suffered from the last 30–40yrs could be perfectly answered through crypto technology and that is why we have added a carbon credit crypto project to the Nonce Classic portfolio. There have been many teams out there that have tried to solve environmental problems through crypto but very few that have measurable experience working in the carbon credit scene. Thus we have put in our efforts to find projects that are not crypto projects created for the sake of issuing tokens but projects that pragmatically use crypto technology to combat climate change by solving problems of the current carbon credit market. In that process, we came to hear of Climate Coin, a veritable carbon credit crypto project, and us Nonce Classic as an accelerator, have begun contributing to its growth and invested in its tokens. Starting with this article, we plan to publish a series of articles explaining why the carbon credit market is bullish, why we invested in Climate Coin, and what kind of project Climate Coin is specifically. In this first article let us understand the carbon credit market and look into its growth potential! Let’s begin :)

The Unavoidable Entry of the Net Zero Era

Net zero means... Human carbon emissions are balanced by carbon reduction efforts. A non-environmentalist may find it hard to accept that net zero is attainable by 2050. Global cooperation to save the earth is happening faster than we imagine.

In the Paris Agreement of COP26, concluded in Glasgow, UK on Oct. 31, 2021, nations pledged to reduce worldwide yearly greenhouse gas emissions by more than 50% by 2030 and attain net zero by 2050. Governments throughout the world have pledged net zero at the national level and are holding each other accountable by submitting Nationally Determined Contributions (NDC) every five years to assess implementation. 127 of 198 nations have declared net zero.

Each country's 1.5-degree reduction plans have led to carbon reduction obligations for companies. In places with the strictest environmental regulations, like the EU, companies often face bankruptcy because the cost of buying carbon credits to meet their carbon allowances exceeds their operating profits. In this day and age, minimizing carbon emissions and securing carbon credits are crucial.

Recent SEC actions on climate change may increase companies' concerns about reducing emissions. The SEC required all U.S. stock market companies to disclose their annual greenhouse gas emissions and climate change impact on March 21, 2022. The SEC prepared the proposed regulation through in-depth analysis and stakeholder input since last year. Three out of four SEC members agreed that it should pass without major changes. If the regulation passes, it will affect not only US companies, but also countless companies around the world, directly or indirectly.

Even companies not listed on the U.S. stock market will be affected and, in most cases, required to disclose emissions. Companies listed on the U.S. stock market with significant greenhouse gas emissions or specific targets are subject to stricter emission standards (Scope 3) and disclosure obligations, which will magnify investigations into all related companies. Greenhouse gas emissions can be calculated three ways. Scope 1 measures carbon emissions from a company's facilities and transportation. Scope 2 measures carbon emissions from energy purchases. Scope 3 covers all indirect emissions from a company's value chains.

The SEC's proposed carbon emission disclosure mandate and regulations are one example of how carbon credit policies can cross borders and affect all parties. As such incidents will continue throughout the implementation of net zero, even companies that are not immediately obligated to disclose their carbon emissions must be prepared to respond to changes in carbon emission laws and policies.

Carbon reduction obligations will soon become individual. Individual consumption has increased dramatically with improved quality of life and convenience, despite national and corporate efforts to reduce carbon emissions. Since consumption is directly related to carbon emissions, increasing consumption increases carbon emissions. Countries around the world have agreed that to achieve net zero, carbon emissions must be reduced on an individual level. Solutions to individual carbon reduction are being actively discussed and studied under the term Personal Carbon Trading (PCT).

PCT is a system that allows individuals to trade carbon emission quotas in the form of carbon credits. Individuals who emit more carbon than their allotment can buy carbon credits from those who emit less. European cities with well-established carbon credit markets are preparing for net zero by conducting early carbon reduction prototype projects. The era of checking product labels for carbon footprints, choosing low-emissions transportation, and worrying about hot shower emissions is closer than we think.

The Market for Carbon Credits Is Expanding Fearfully

Compliance and voluntary carbon markets make up the carbon credit market.

A Compliance Market enforces carbon emission allowances for actors. Companies in industries that previously emitted a lot of carbon are included in the mandatory carbon market, and each government receives carbon credits each year. If a company's emissions are less than the assigned cap and it has extra carbon credits, it can sell them to other companies that have larger emissions and require them (Cap and Trade). The annual number of free emission permits provided to companies is designed to decline, therefore companies' desire for carbon credits will increase. The compliance market's yearly trading volume will exceed $261B in 2020, five times its 2017 level.

In the Voluntary Market, carbon reduction is voluntary and carbon credits are sold for personal reasons or to build market participants' eco-friendly reputations. Even if not in the compliance market, it is typical for a corporation to be obliged to offset its carbon emissions by acquiring voluntary carbon credits. When a company seeks government or company investment, it may be denied because it is not net zero. If a significant shareholder declares net zero, the companies below it must execute it. As the world moves toward ESG management, becoming an eco-friendly company is no longer a strategic choice to gain a competitive edge, but an important precaution to not fall behind. Due to this eco-friendly trend, the annual market volume of voluntary emission credits will approach $1B by November 2021. The voluntary credit market is anticipated to reach $5B to $50B by 2030. (TSCVM 2021 Report)

In conclusion

This article analyzed how net zero, a target promised by countries around the world to combat climate change, has brought governmental, corporate, and human changes. We discussed how these shifts will become more obvious as we approach net zero, and how the carbon credit market would increase exponentially in response. In the following piece, let's analyze the hurdles impeding the carbon credit market's growth, how the project we invested in tries to tackle these issues, and why we chose Climate Coin. Wait! Jim Skea, co-chair of the IPCC working group, said,

“It’s now or never, if we want to limit global warming to 1.5°C” — Jim Skea

Join nonceClassic’s community:

Telegram: https://t.me/non_stock

Youtube: https://www.youtube.com/channel/UCqeaLwkZbEfsX35xhnLU2VA

Twitter: @nonceclassic

Mail us : general@nonceclassic.org

ERC165 (or EIP-165) is a standard utilized by various open-source smart contracts like Open Zeppelin or Aavegotchi.

What's it? You must implement? Why do we need it? I'll describe the standard and answer any queries.

What is ERC165

ERC165 detects and publishes smart contract interfaces. Meaning? It standardizes how interfaces are recognized, how to detect if they implement ERC165, and how a contract publishes the interfaces it implements. How does it work?

Why use ERC165? Sometimes it's useful to know which interfaces a contract implements, and which version.

Identifying interfaces

An interface function's selector. This verifies an ABI function. XORing all function selectors defines an interface in this standard. The following code demonstrates.

// SPDX-License-Identifier: UNLICENCED
pragma solidity >=0.8.0 <0.9.0;

interface Solidity101 {
    function hello() external pure;
    function world(int) external pure;
}

contract Selector {
    function calculateSelector() public pure returns (bytes4) {
        Solidity101 i;
        return i.hello.selector ^ i.world.selector;
        // Returns 0xc6be8b58
    }

    function getHelloSelector() public pure returns (bytes4) {
        Solidity101 i;
        return i.hello.selector;
        // Returns 0x19ff1d21
    }

    function getWorldSelector() public pure returns (bytes4) {
        Solidity101 i;
        return i.world.selector;
        // Returns 0xdf419679
    }
}

This code isn't necessary to understand function selectors and how an interface's selector can be determined from the functions it implements.

Run that sample in Remix to see how interface function modifications affect contract function output.

Contracts publish their implemented interfaces.

We can identify interfaces. Now we must disclose the interfaces we're implementing. First, import IERC165 like so.

pragma solidity ^0.4.20;

interface ERC165 {
    /// @notice Query if a contract implements an interface
    /// @param interfaceID The interface identifier, as specified in ERC-165
    /// @dev Interface identification is specified in ERC-165. 
    /// @return `true` if the contract implements `interfaceID` and
    ///  `interfaceID` is not 0xffffffff, `false` otherwise
    function supportsInterface(bytes4 interfaceID) external view returns (bool);
}

We still need to build this interface in our smart contract. ERC721 from OpenZeppelin is a good example.

// SPDX-License-Identifier: MIT
// OpenZeppelin Contracts (last updated v4.5.0) (token/ERC721/ERC721.sol)

pragma solidity ^0.8.0;

import "./IERC721.sol";
import "./extensions/IERC721Metadata.sol";
import "../../utils/introspection/ERC165.sol";
// ...

contract ERC721 is Context, ERC165, IERC721, IERC721Metadata {
  // ...

  function supportsInterface(bytes4 interfaceId) public view virtual override(ERC165, IERC165) returns (bool) {
    return
      interfaceId == type(IERC721).interfaceId ||
      interfaceId == type(IERC721Metadata).interfaceId ||
      super.supportsInterface(interfaceId);
  }
  
  // ...
}

I deleted unnecessary code. The smart contract imports ERC165, IERC721 and IERC721Metadata. The is keyword at smart contract declaration implements all three.

Kind (interface).

Note that type(interface).interfaceId returns the same as the interface selector.

We override supportsInterface in the smart contract to return a boolean that checks if interfaceId is the same as one of the implemented contracts.

Super.supportsInterface() calls ERC165 code. Checks if interfaceId is IERC165.

function supportsInterface(bytes4 interfaceId) public view virtual override returns (bool) {
    return interfaceId == type(IERC165).interfaceId;
}

So, if we run supportsInterface with an interfaceId, our contract function returns true if it's implemented and false otherwise. True for IERC721, IERC721Metadata, andIERC165.

Conclusion

I hope this post has helped you understand and use ERC165 and why it's employed.

Have a great day, thanks for reading!

Homeownership isn't the only life pattern.

Want to irritate people?

My party trick is to say I used to own a house but no longer do.

I no longer wish to own a home, not because I lost it or because I'm moving.

It was a long-term plan. It was more deliberate than buying a home. Many people are committed for this reason.

Poppycock.

Anyone who told me that owning a house (or striving to do so) is a must is wrong.

Because, URGH.

One pattern for life is to own a home, but there are millions of others.

You can afford to buy a home? Go, buddy.

You think you need 1,000 square feet (or more)? You think it's non-negotiable in life?

Nope.

It's insane that society forces everyone to own real estate, regardless of income, wants, requirements, or situation. As if this trade brings happiness, stability, and contentment.

Take it from someone who thought this for years: drywall isn't happy. Living your way brings contentment.

That's in real estate. It may also be renting a small apartment in a city that makes your soul sing, but you can't afford the downpayment or mortgage payments.

Living or traveling abroad is difficult when your life savings are connected to something that eats your money the moment you sign.

#vanlife, which seems like torment to me, makes some people feel alive.

I've seen co-living, vacation rental after holiday rental, living with family, and more work.

Insisting that home ownership is the only path in life is foolish and reduces alternative options.

How little we question homeownership is a disgrace.

No one challenges a homebuyer's motives. We congratulate them, then that's it.

When you offload one, you must answer every question, even if you have a loose screw.

• Why do you want to sell?

• Do you have any concerns about leaving the market?

• Why would you want to renounce what everyone strives for?

• Why would you want to abandon a beautiful place like that?

• Why would you mismanage your cash in such a way?

• But surely it's only temporary? RIGHT??

Incorrect questions. Buying a property requires several inquiries.

• The typical American has $4500 saved up. When something goes wrong with the house (not if, it’s never if), can you actually afford the repairs?

• Are you certain that you can examine a home in less than 15 minutes before committing to buying it outright and promising to pay more than twice the asking price on a 30-year 7% mortgage?

• Are you certain you're ready to leave behind friends, family, and the services you depend on in order to acquire something?

• Have you thought about the connotation that moving to a suburb, which more than half of Americans do, means you will be dependent on a car for the rest of your life?

Plus:

Are you sure you want to prioritize home ownership over debt, employment, travel, raising kids, and daily routines?

Homeownership entails that. This ex-homeowner says it will rule your life from the time you put the key in the door.

This isn't questioned. We don't question enough. The holy home-ownership grail was set long ago, and we don't challenge it.

Many people question after signing the deeds. 70% of homeowners had at least one regret about buying a property, including the expense.

Exactly. Tragic.

Homes are different from houses

We've been fooled into thinking home ownership will make us happy.

Some may agree. No one.

Bricks and brick hindered me from living the version of my life that made me most comfortable, happy, and steady.

I'm spending the next month in a modest apartment in southern Spain. Even though it's late November, today will be 68 degrees. My spouse and I will soon meet his visiting parents. We'll visit a Sherry store. We'll eat, nap, walk, and drink Sherry. Writing. Jerez means flamenco.

That's my home. This is such a privilege. Living a fulfilling life brings me the contentment that buying a home never did.

I'm happy and comfortable knowing I can make almost all of my days good. Rejecting home ownership is partly to blame.

I'm broke like most folks. I had to choose between home ownership and comfort. I said, I didn't find them together.

Feeling at home trumps owning brick-and-mortar every day.

The following is the reality of what it's like to turn the American Dream around.

Leaving the housing market.

Sometimes I wish I owned a home.

I miss having my own yard and bed. My kitchen, cookbooks, and pizza oven are missed.

But I rarely do.

Someone else's life plan pushed home ownership on me. I'm grateful I figured it out at 35. Many take much longer, and some never understand homeownership stinks (for them).

It's confusing. People will think you're dumb or suicidal.

If you read what I write, you'll know. You'll realize that all you've done is choose to live intentionally. Find a home beyond four walls and a picket fence.

Miss? As I said, they're not home. If it were, a pizza oven, a good mattress, and a well-stocked kitchen would bring happiness.

No.

If you can afford a house and desire one, more power to you.

There are other ways to discover home. Find calm and happiness. For fun.

For it, look deeper than your home's foundation.

Quitting your job for a side gig isn't smart.

A few years ago, I would have laughed at the idea of starting a side business.

I never thought a side gig could earn more than my 9-to-5. My side gig pays more than my main job now.

You may then tell me to leave your job.  But I don't want to gamble, and my side gig is important. Programming and web development help me write better because of my job.

Yes, I share work-related knowledge. Web development, web3, programming, money, investment, and side hustles are key.

Let me now show you how to make one.

Create a side business based on your profession or your interests.

I'd be direct.

Most people don't know where to start or which side business to pursue.

You can make money by taking online surveys, starting a YouTube channel, or playing web3 games, according to several blogs.

You won't make enough money and will waste time.

Nitin directs our efforts. My friend, you've worked and have talent. Profit from your talent.

Example:

College taught me web development. I soon created websites, freelanced, and made money. First year was hardest for me financially and personally.

As I worked, I became more skilled. Soon after, I got more work, wrote about web development on Medium, and started selling products.

I've built multiple income streams from web development. It wasn't easy. Web development skills got me a 9-to-5 job.

Focus on a specific skill and earn money in many ways. Most people start with something they hate or are bad at; the rest is predictable.

Result? They give up, frustrated.

Quietly focus for a year.

I started my side business in college and never told anyone. My parents didn't know what I did for fun.

The only motivation is time constraints. So I focused.

As I've said, I focused on my strengths (learned skills) and made money. Yes, I was among Medium's top 500 authors in a year and got a bonus.

How did I succeed? Since I know success takes time, I never imagined making enough money in a month. I spent a year concentrating.

I became wealthy. Now that I have multiple income sources, some businesses pay me based on my skill.

I recommend learning skills and working quietly for a year. You can do anything with this.

The hardest part will always be the beginning.

When someone says you can make more money working four hours a week. Leave that, it's bad advice.

If someone recommends a paid course to help you succeed, think twice.

The beginning is always the hardest.

I made many mistakes learning web development. When I started my technical content side gig, it was tough. I made mistakes and changed how I create content, which helped.

And it’s applicable everywhere.

Don't worry if you face problems at first. Time and effort heal all wounds.

Quitting your job to work a side job is not a good idea.

Some honest opinions.

Most online gurus encourage side businesses. It takes time to start and grow a side business.

Suppose you quit and started a side business.

After six months, what happens? Your side business won't provide enough money to survive.

Indeed. Later, you'll become demotivated and tense and look for work.

Instead, work 9-5, and start a side business. You decide. Stop watching Netflix and focus on your side business.

I know you're busy, but do it.

Next? It'll succeed or fail in six months. You can continue your side gig for another six months because you have a job and have tried it.

You'll probably make money, but you may need to change your side gig.

That’s it.

You've created a new revenue stream.

Remember.

Starting a side business, a company, or finding work is difficult. There's no free money in a competitive world. You'll only succeed with skill.

Read it again.

Focusing silently for a year can help you succeed.

I studied web development and wrote about it. First year was tough. I went viral, hit the top 500, and other firms asked me to write for them. So, my life changed.

Yours can too. One year of silence is required.

Enjoy!

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.