“Perfectionism is the enemy of the idea muscle. " — James Altucher

Writer's block is a typical explanation for low output. Success requires productivity.

In four years of writing, I've never had writer's block. And you shouldn't care.

You'll never run out of content ideas if you follow a few tactics. No, I'm not overpromising.

Take Note of Ideas

Brains are strange machines. Blank when it's time to write. Idiot. Nothing. We get the best article ideas when we're away from our workstation.

• In the shower

• Driving

• In our dreams

• Walking

• During dull chats

• Meditating

• In the gym

No accident. The best ideas come in the shower, in nature, or while exercising.

(Your workstation is the worst place for creativity.)

The brain has time and space to link 'dots' of information during rest. It's eureka! New idea.

If you're serious about writing, capture thoughts as they come.

Immediately write down a new thought. Capture it. Don't miss it. Your future self will thank you.

As a writer, entrepreneur, or creative, letting ideas slide is bad.

I recommend using Evernote, Notion, or your device's basic note-taking tool to capture article ideas.

It doesn't matter whatever app you use as long as you collect article ideas.

When you practice 'idea-capturing' enough, you'll have an unending list of article ideas when writer's block hits.

High-Quality Content

More books, films, Medium pieces, and Youtube videos I consume, the more I'm inspired to write.

What you eat shapes who you are.

Celebrity gossip and fear-mongering news won't help your writing. It won't help you write regularly.

Instead, read expert-written books. Watch documentaries to improve your worldview. Follow amazing people online.

Develop your 'idea muscle' Daily creativity takes practice. The more you exercise your 'idea muscles,' the easier it is to generate article ideas.

I've trained my 'concept muscle' using James Altucher's exercise.

Write 10 ideas daily.

Write ten book ideas every day if you're an author. Write down 10 business ideas per day if you're an entrepreneur. Write down 10 investing ideas per day.

Write 10 article ideas per day. You become a content machine.

It doesn't state you need ten amazing ideas. You don't need 10 ideas. Ten ideas, regardless of quality.

Like at the gym, reps are what matter. With each article idea, you gain creativity. Writer's block is no match for this workout.

Quit Perfectionism

Perfectionism is bad for writers. You'll have bad articles. You'll have bad ideas. OK. It's creative.

Writing success requires prolificacy. You can't have 'perfect' articles.

“Perfectionism is the enemy of the idea muscle. Perfectionism is your brain trying to protect you from harm.” — James Altucher

Vincent van Gogh painted 900 pieces. The Starry Night is the most famous.

Thomas Edison invented 1093 things, but not all were as important as the lightbulb or the first movie camera.

Mozart composed nearly 600 compositions, but only Serenade No13 became popular.

Always do your best. Perfectionism shouldn't stop you from working. Write! Publicize. Make. Even if imperfect.

Write Your Story

Living an interesting life gives you plenty to write about. If you travel a lot, share your stories or lessons learned.

Describe your business's successes and shortcomings.

Share your experiences with difficulties or addictions.

More experiences equal more writing material.

If you stay indoors, perusing social media, you won't be inspired to write.

Have fun. Travel. Strive. Build a business. Be bold. Live a life worth writing about, and you won't run out of material.

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.

The company seeking 24/7 ultra-powerful solar electricity.

We're rushing to adopt low-carbon energy to prevent a self-made doomsday. We're using solar, wind, and wave energy. These low-carbon sources aren't perfect. They consume large areas of land, causing habitat loss. They don't produce power reliably, necessitating large grid-level batteries, an environmental nightmare. We can and must do better than fossil fuels. Longi, one of the world's top solar panel producers, is creating a low-carbon energy source. Solar-powered spacecraft. But how does it work? Why is it so environmentally harmonious? And how can Longi unlock it?

Space-based solar makes sense. Satellites above Medium Earth Orbit (MEO) enjoy 24/7 daylight. Outer space has no atmosphere or ozone layer to block the Sun's high-energy UV radiation. Solar panels can create more energy in space than on Earth due to these two factors. Solar panels in orbit can create 40 times more power than those on Earth, according to estimates.

How can we utilize this immense power? Launch a geostationary satellite with solar panels, then beam power to Earth. Such a technology could be our most eco-friendly energy source. (Better than fusion power!) How?

Solar panels create more energy in space, as I've said. Solar panel manufacture and grid batteries emit the most carbon. This indicates that a space-solar farm's carbon footprint (which doesn't need a battery because it's a constant power source) might be over 40 times smaller than a terrestrial one. Combine that with carbon-neutral launch vehicles like Starship, and you have a low-carbon power source. Solar power has one of the lowest emissions per kWh at 6g/kWh, so space-based solar could approach net-zero emissions.

Space solar is versatile because it doesn't require enormous infrastructure. A space-solar farm could power New York and Dallas with the same efficiency, without cables. The satellite will transmit power to a nearby terminal. This allows an energy system to evolve and adapt as the society it powers changes. Building and maintaining infrastructure can be carbon-intensive, thus less infrastructure means less emissions.

Space-based solar doesn't destroy habitats, either. Solar and wind power can be engineered to reduce habitat loss, but they still harm ecosystems, which must be restored. Space solar requires almost no land, therefore it's easier on Mother Nature.

Space solar power could be the ultimate energy source. So why haven’t we done it yet?

Well, for two reasons: the cost of launch and the efficiency of wireless energy transmission.

Advances in rocket construction and reusable rocket technology have lowered orbital launch costs. In the early 2000s, the Space Shuttle cost $60,000 per kg launched into LEO, but a SpaceX Falcon 9 costs only $3,205. 95% drop! Even at these low prices, launching a space-based solar farm is commercially questionable.

Energy transmission efficiency is half of its commercial viability. Space-based solar farms must be in geostationary orbit to get 24/7 daylight, 22,300 miles above Earth's surface. It's a long way to wirelessly transmit energy. Most laser and microwave systems are below 20% efficient.

Space-based solar power is uneconomical due to low efficiency and high deployment costs.

Longi wants to create this ultimate power. But how?

They'll send solar panels into space to develop space-based solar power that can be beamed to Earth. This mission will help them design solar panels tough enough for space while remaining efficient.

Longi is a Chinese company, and China's space program and universities are developing space-based solar power and seeking commercial partners. Xidian University has built a 98%-efficient microwave-based wireless energy transmission system for space-based solar power. The Long March 5B is China's super-cheap (but not carbon-offset) launch vehicle.

Longi fills the gap. They have the commercial know-how and ability to build solar satellites and terrestrial terminals at scale. Universities and the Chinese government have transmission technology and low-cost launch vehicles to launch this technology.

It may take a decade to develop and refine this energy solution. This could spark a clean energy revolution. Once operational, Longi and the Chinese government could offer the world a flexible, environmentally friendly, rapidly deployable energy source.

Should the world adopt this technology and let China control its energy? I'm not very political, so you decide. This seems to be the beginning of tapping into this planet-saving energy source. Forget fusion reactors. Carbon-neutral energy is coming soon.

This is how to build a car while driving.

Building a high-growth startup is compared to building a car while it's speeding down the highway.

How to plan without going crazy? Or, without losing team, board, and investor buy-in?

I just delivered our company's product roadmap for the rest of the year. Complete. Thorough. Page-long. I'm optimistic about its chances of surviving as everything around us changes, from internal priorities to the global economy.

It's tricky. This isn't the first time I've created a startup roadmap. I didn't invent a document. It took time to deliver a document that will be relevant for months.

Goals matter.

Although they never change, goals are rarely understood.

This is the third in a series about a startup's unique roadmapping needs. Velocity is the intensity at which a startup must produce to survive.

A high-growth startup moves at breakneck speed, which I alluded to when I said priorities and economic factors can change daily or weekly.

At that speed, a startup's roadmap must be flexible, bend but not break, and be brief and to the point. I can't tell you how many startups and large companies develop a product roadmap every quarter and then tuck it away.

Big, wealthy companies can do this. It's suicide for a startup.

The drawer thing happens because startup product roadmaps are often valid for a short time. The roadmap is a random list of features prioritized by different company factions and unrelated to company goals.

It's not because the goals changed that a roadmap is shelved or ignored. Because the company's goals were never communicated or documented in the context of its product.

In the previous post, I discussed how to turn company goals into a product roadmap. In this post, I'll show you how to make a one-page startup roadmap.

In a future post, I'll show you how to follow this roadmap. This roadmap helps you track company goals, something a roadmap must do.

Be vague for growth, but direct for execution.

Here's my plan. The real one has more entries and more content in each.

Let's discuss smaller boxes.

Product developers and engineers know that the further out they predict, the more wrong they'll be. When developing the product roadmap, this rule is ignored. Then it bites us three, six, or nine months later when we haven't even started.

Why do we put everything in a product roadmap like a project plan?

Yes, I know. We use it when the product roadmap isn't goal-based.

A goal-based roadmap begins with a document that outlines each goal's idea, execution, growth, and refinement.

Once the goals are broken down into epics, initiatives, projects, and programs, only the idea and execution phases should be modeled. Any goal growth or refinement items should be vague and loosely mapped.

Why? First, any idea or execution-phase goal will result in growth initiatives that are unimaginable today. Second, internal priorities and external factors will change, but the goals won't. Locking items into calendar slots reduces flexibility and forces deviation from the single source of truth.

No soothsayers. Predicting the future is pointless; just prepare.

A map is useless if you don't know where you're going.

As we speed down the road, the car and the road will change. Goals define the destination.

This quarter and next quarter's roadmap should be set. After that, you should track destination milestones, not how to get there.

When you do that, even the most critical investors will understand the roadmap and buy in. When you track progress at the end of the quarter and revise your roadmap, the destination won't change.