A Progressive Journey Through Working With AI Art – Part 4 – Better Prompts

The next step in my journey to better AI Art, was better prompts. Which also has sort of landed me on just using one complex prompt I found and modifying it as needed, which works very well. I started off by adding more descriptive words to the basic prompts. Including Camera models which was suggested by quite a few people.

  • “In the Style of Manga”
  • “An oil Painting Of”
  • “A Pencil Sketch of”
  • “in the style of [artist]
  • “Realistic”
  • “Hyper-realistic”
  • Canon 5D

This worked better. But I started looking around on the Stable diffusion Sub-Reddits for good prompts to use. I came across the following Prompt:

, (humorous illustration, hyperrealistic, big depth of field, colors, night club scenery, 3d octane render, 4k, concept art, hyperdetailed, hyperrealistic, trending on artstation:1.1)

Negatives:
text, b&w, (cartoon, 3d, bad art, poorly drawn, close up, blurry, disfigured, deformed, extra limbs:1.5)

Which I have used and adapted quite a lot. Essentially, everything in front of the first Comma is your actual prompt. This is essentially, what I have been doing. Everything after refines things a lot. You can also change the background by editing the “night club scenery” bit.

Anyway, the rest of the post is sharing some more pics based on this prompt.

Prompt: “Tracer from Overwatch” +

As normal, really iffy on the hands, but still some neat concepts that could actually be skins in the game.

Prompt: Godzilla +

Prompt: Several different Batman Prompts (Batman Fighting, Batman Overlooking Gotham, Batman Battling Joker)

Prompt: The Joker +

These are some of my favorites so far. I am not a huge Joker Fan really, but they do a REALLY good job of portraying the more modern crazy that is The Joker. I actually left a few off because frankly, they are super creepy, but really are nice.

Prompt: Professor Layton

Again, it has no idea who Layton is, but still seems to do really well with the Aethetic of Layton. Which is kind of odd honestly.

Prompt: An Adorable Pixar Kitten

Feels like Pixar styled art is cheater mode a bit but these came out pretty good as well.

Three Prompts with similar results, A Norwegian Landscape, The Lord of the Rings, and Arya Stark,

It’s kind of crazy just how much better the results have gotten from previous attempts, especially just like, 6 months ago or something, when I started playing with this concept using online tools. That said, it also gets old pretty quick, and you end up with a lot of “Weird shit” output, extra limbs, weird proportions, extra elbows, odd faces. I can see how it might be useful to produce some generic banner backdrops and whatnot. I also can see it just getting even better, very rapidly. If hands can be figured out, that would be a real game changer.

A Progressive Journey Through Working With AI Art – Part 3 – Running Natively with Automatic1111

After experimenting with online sources, then running Stable Diffusion locally using Windows Subsystem for Linux, I wanted more, and better, because I knew my machine was capable of much more. So I looked into alternatives and found Automatic1111’s Stable Diffusion variant.

The core take away here, is how this is like night and day in performance and quality.

Previously, with WSL, I would run batches of prompts and seeds and maybe get a few okish results. Also, any dimensions larger than the base 512×512 would crash the thing and I’d get nothing. Basically, it definitely was not exploiting the full potential here. It also completely dogged my entire rig down while building an image, which took maybe 5-10 minutes for the actual processing to work.

It still dogs the machine down, but not nearly as much as it had been. And it takes like 10-20 SECONDS to produce an image. The image quality is also like 1000 times better, though not still without that “AI Art Wonkeyness” like these weird square cows.

—- Read More —-: A Progressive Journey Through Working With AI Art – Part 3 – Running Natively with Automatic1111 Read More

100 Days of Python, Projects 71-80 (but not really) #100DaysofCode

Ok, I’m going to be frank here.

I’m skipping most of the Data Analytics portion of this course.  It’s 9 Lessons, Day 71-80.  It is all done using Google Colab Notebooks, it’s all extremely, extremely, extremely, repetitive, yet I am not really feeling like I am learning anything.

Every lesson amounts to, Open this Google Collab Notebook, which is a new and… interesting tool, I guess it’s similar to a Jupyter Notebook.  It’s basically, running “code” in weird choppy step by step chunks.  Each lesson amounts to.

  • Open a provided CSV file.
  • Make a Graph
  • Format the Graph.
  • Maybe do some layered graphs.
  • Sometimes the type of graph is different (line, bar, scatter, etc).

It’s dull, it’s repetitive, I have almost zero interest in statistics and data science honestly.  Fun Fact, Statistics is the one class I dropped in college.  I could have gotten a minor in Math with my Mechanical Engineering Degree with one more math credit, so I took Statistics.  It was the only class I had any actual homework in my last semester, everything else was writing reports, the homework was absolutely brutal, and the entire class felt like butchering math to create Conformation Bias.  It was stupid, it doesn’t account for nuance and one offs, which is what you should actually care about because those are the failure points.  I could go on and on.

Anyway, the sections tarted out pretty fun, learning  new tool, making fancy graphs, then halfway through I found I was just copying and pasting answers from the lecture because I just wanted it to be over and I didn’t care about the material.  

So I’m just, skipping it. This isn’t a graded University Course, hell I have made plenty of small side projects during this course to fill in the idea of “100 Projects in 100 Days”.

Also, if I wanted to make pretty charts, I would just use Microsoft Excel.  Because I am already pretty good at that.

This leads into the final “stretch” of this course, which I have already started working on some.  The Final 19 Projects, which are all basically, open ended suggestions.  There is a Data Analytics one in there, which I actually probably will go ahead and throw together some Charts for.  I have the perfect Data Set for it, ever since I bought my car, back in 2015, I’ve tracked my mileage and fuel usage every time I fill up (in Excel).  I can dump out a huge CSV was and manipulate that data in Google Collab or a Jupyter Notebook.

A Progressive Journey Through Working With AI Art – Part 2 – Running Locally

My initial foray into doing AI Art locally involved Stable Diffusion, which can be found here. This path ended up being sort of, multi staged. Initially I set it all up to run off the command line through PowerShell and Python. I ran through several prompts aver a few days off and on with iffy results (above). At some point I closed it down. When I came back, I couldn’t get it to run again. I couldn’t figure out WHY and so I gave up and nuked it to start fresh. (The problem was I forgot to use the Python Virtual Environment, DUUUUH).

Part 1 can be found here.

Side Note, Unfortunately, I don’t know the prompts for many of these because my initial runs just produced file names with numbers for file names.

—- Read More —-: A Progressive Journey Through Working With AI Art – Part 2 – Running Locally Read More

100 Days of Python, Projects 66-70 #100DaysofCode

Whew, I didn’t really think I’d get to 9 parts in this series, and I am only around 2/3rds of the way through even.  I actually may change up the format later with the last 20 projects that are listed as “Professional”.  Maybe one post each.

The bulk of this round is wrapping up the Flask projects and building a simple blog that runs on Python.  It’s been fun.  I’ve been a bit busier than normal slow my pace has slowed, but that’s ok too.  Day 66 in particular felt like it took longer than it really should have, given how little it felt like it was doing.

Day 66 – Build RESTful APIs

Kind of a different sort of project.  It’s building something, but not really anything, with any sort of interface.  All of the interaction is done through Postman (or the URL if one wants), and the responses are all JSON of some sort.  

We built 6 API interfaces to work with a database of Coffee shops.

  • One returned all of the shops.
  • One returned all of them in a particular city.
  • One returned a specific shop only.
  • One updated the price of coffee.
  • One added a new Coffee Shop.
  • One deleted a Coffee Shop.

Including built in error handling for if a Coffee Shop didn’t exist etc.  It all feels like it could be useful later in the Blog Project.  Not too useful on it’s own accord.

Day 67 – Blog Capstone Project Part 3

Back to the Blog Project, and after this round it’s a LOT more Blog Like, though incredibly insecure.  The Security Part looks to be the subject of the next couple of lessons though, so it’s all good.  We basically started with the last Blog Project from Day 59, and added the ability to add, edit, and delete posts.  Also with a Database back end.

One tricky moment here.  The idea was to use the same form page for New Post and Edit Post.  Originally I had nested these together, and had some if/else cases to check if it was Editing or Adding, and kind of got stuck on how to handle the date, since the date doesn’t get changed on updates.

But then I had a bit of an epiphany moment, the kind you get often with coding, where you band against a problem, convinced it’s the way to solve things, only to realize there is an obvious, easy method.

I split them up.  No more if/else, and no more worrying about what the DB is doing at the end.  Because two separate @app.route and related functions, can route to the same HTML template (the Edit Form).  The only thing needed was to pass a header variable along to change the title.  

It was a total “Duuuuuuh” moment.

Anyway, no security though, because there are no users or security keys and anyone can just come in and post anything and delete anything and it would be total chaos as an actual production Blog platform.

Day 68 – Authentication with Flask

Oops, I spoiled what this topic was going to be in the last project write up.  This whole section is working with Authentication using a Database for persistence.  It’s a simple website, a home page, a user creation page, a log in page, and a secrets page.  Users can create an account, and once registered, they can access the Secrets page and download a PDF.  

Part of this exercise is also restricting access for not logged in users and another part is proper security and handling passwords with hashing and salting.

Of all of the projects so far, I feel like this may be one of the most important ones, despite it’s basic simplicity.  I’ve been planning to work out the best way to share some of the projects built so far, and using Flask was a good “first step”, but being security conscious, I would rather not throw a bunch of web based Flask pages up with no restrictions on access.  Like last session, there was the Library App or the Top Ten Movies page.  Sure, I could put them up on the web-server, map some sub domain to them, but anyone can edit them, without proper, persistent security and restricted access.  

Now, I can make that happen.  Plus, with the modulator of routing and such, I can easily slip them into parts of the overall Blog Project, in time.

Day 69 – Blog Capstone Project Part 4

And here it is, the culmination of the Python based Flask Blog.  It’s neat.  I like it, well, I like the basic functionality of it.  The layout is a bit plain but that’s fixed.  I doubt it ever replaces WordPress, I love WordPress, but I could definitely see uses for this finished product.  I may actually modify it to work as a sort of “Twitter Replacement” since Twitter is currently burning down.  Using what I have learned, I could easily set this up to take a post, then “Syndicate” it out via API calls to Twitter, Facebook, Mastodon, or anywhere.  While keeping my own archive of posts.

My current, next up To Do Items:

  • Create RSS Feed Page  
  • Create Admin Page  
  • Enable/Disable Comments for Posts  
  • Allow Admin to delete Comments  
  • Allow Users to delete their own comments  
  • Add Mailer.class and add email notifications  
  • Add Pagination to Home Page  
  • Add Tags and Category Options to Posts

Also, on a side note, this actually isn’t the first time I’ve built a “Blog Platform”. I built a basic one a few years ago in HTML and PHP. Heck, the system I used int he early 2000s with SHTML Pages was sort of a “Blog Platform”.

I’d recommend, for anyone going through this course, check the notes on the Blog courses.  There are a lot of good suggestions for improvements, especially in stopping things like Java Script injection in the comments.  When you go to Angela’s (The Instructor) Blog link, it immediately throws out a JavaScript pop up that someone dropped in the comments.

Day 70 – Hosting with Heroku

So, there isn’t really a Day 70 project.  I did go through the section, and the most useful part for me was the more robust SQL solution mentioned in the last section.  The first few parts were about GIT and Github, which I am already using.  The middle part was about using Heroku, which I have heard of and used a bit before but for the long term, I don’t need to use a freemium hosting service in a jankey way.  I have a VPS, and later I will figure out how to get Apache to play nice with Flask.

For now, I’ve split the Blog Project into it’s own repository and worked up my initial planned ToDo List.  But I need to keep my focus on the course for now.