Metapost | 2018.05.31

Deep learning neural style transfer DL4J
New meta

Sometimes there are fun conversations I think about transcribing, so I added support in my web tools. This, of course, meant partly manifesting a longstanding desire to create avatars for everyone mentioned herein. The second part is ongoing, the first part took an hour or two at the keyboard.

me
me
... until Alf gets caught roofiing other castmembers.
is that how you spell it? sants
sants
I wouldn't know. But I assume it's like skiing.
or snookiing
It's actually Zac that engages your mindless celebrity banter. I'm the one you talk to about VPNFilter. And Reverse Scrum.
wtf is reverse scrum?
It's like Scrum, but reverse. It's the hottest new thing in software process engineering.
You write the code, then tell people how long it took you and then figure out why you did it?
It's still considered scrum because there's no testing.
me
me
Dang, this guy is ready to be an uncertified ScrumSlave. You just need to take my six week course.
I am a certified Scrum Master from a previous job, actually. Somewhere there's a printed piece of paper that confirms it. sants
sants
Do you ever wonder about your life choices?
No need. I'm doing life in reverse Scrum. By the time I need to figure out why I did it, I'll be dead.
That's the inagility that makes a great engineer! Don't forget to determine how long things took by playing a dice game.
That makes me more agile actually. If there is no plan, it's impossible to be inagile.
There is a plan, it just happens in reverse. Good god I need to buy you a Reverse Scrum book.
Here's your super meta certificate:

[Ed: my markup parser has been updated since originally posting.]
Soirees

Battleship board game

Jes and I have had some casual evenings out. In one I got wrecked at Battleship. And Scrabble.

Birch Scripps aquarium tank shark fish kelp Campari

We also went to a concert at Birch.
Style transfer

Deep learning DL4J style transfer Zero motorcycle cafe racer art

Not a ton of work on the neural side of things, but did get larger images going along with some code refactor.

Deep learning neural style transfer DL4J Picasso
LED headlight

Deep learning DL4J style transfer Zero motorcycle

Grabbed a Sealight 9003 led lamp for the Zero. Seems wise to go led with an electric vehicle. The swap went well, I was even able to keep the dust boot in the configuration. Of course working with connectors in tight areas is never a quick process.

CRG bar end mirror adapter

Also picked up the CRG in-bar adapter to make the bar end mirror from the Duc work.
Sea of Thieves

Sea of Thieves bombarding ship in port

The lolbaters crew is going fully against the pirate code - lighting up ships in port, ignoring signalling lights, and other piratey things. Rarecrosoft released a minor content update, hoping more comes soon.

Sea of Thieves island view Sea of Thieves ship kraken tentacle



Infopost | 2018.05.13

California capitol building Sacramento

Source.

Here we are for another round of "Who's on the ballot?"

Peter Y. Liu | R

California primary voter guide candidate Peter Liu

#cesp5
[Entire statement]

#pass

Christopher N. Carlson | G

California primary 2018 voter information guide Christopher Carlson

"Teach your children calculus And keep the plant safe Or feathered stones and empty bowls Will also be their fate"
[Entire statement]

Deep.

Johnny Wattenburg | NP

California 2018 primary voter information guide Johnny Wattenburg

Why not!
[Entire statement]

This guy gets it. But no.

Erin Cruz | R

California primary voter guide candidate Erin Cruz

Less Tax, Less, Regulation, Less Government. Together we will Make California Golden Again.
[Entire statement]

Less capitalization and incorrect punctuation please!

Don J. Grundmann | NP

California primary voter guide candidate Don Grundmann

A campaign for Sanity. There's no such thing as "transgender." It does not exist. What does exist are broken people... Go to TheyAreAttackingTheChildren.org to join the Constitution Party - the last and only party in California which will fight to defend our children and families. USA vs Sodom Inc.
[Abridged]

But your email is stoptheirs@hotmail.com. Which is your single issue? Taxation = theft or USA vs the corporation of Sodom?

John "Jack" Crew | R

California primary voter information guide John Crew

Atrocity of abortion-on-demand must end.
[Entire statement]

Too busy for a definite article. I like this guy.

Herbert G. Peters | D

California primary voter guide candidate Herbert Peters

Deomocrat. Christian. Pro-Life. End Inflation. Constitutional Tax. Restore Prosperity: our first 100 years "we the people" became the greatest wealth producers in the world with no minimum wage and after Civil war tax rate of 3.1%, (today over 20%). As government grows...
[Abridged]

Wait, hold on, Herb. I know you've been around a while, but you sound like you missed the democrats going from being the slaveowning party of the South to being the liberal party. Wait...


Oh right! I remember you. Speaking of, another blast from the past:

Ling Ling Shi | NP

California primary voter information guide Ling Ling Shi

Run for God's Heart and America's Freedom, Challenge 2016 U.S. Presidential Election results: 2016 U.S. Presidential Election should be nullified by the U.S. Supreme Court and U.S. should re-hold 45th U.S. Presidential Election; Challenge 10 giant chaos in U.S. Economy and Economy-Related sectors.
[Entire statement]

Still on that 10 giant chaos thing eh? I'm not sure about you as a U.S. Senate Candidate: your qualifications for U.S. Senate Candidate seem suspect considering how U.S. Senate Candidate Ling Ling Shi constructs sentences.

And it looks like Jason M. Hanania has gone from "01100101" to a full paragraph of tech vomit including, of course, blockchain. But we also have a new challenger that's part-Massie Munroe, part-Jason Hanania:

Kevin Mottus | R

California primary voter guide candidate Kevin Mottus

The FCC forced roll out of the new 5G wireless technology and Internet of THings will force a cell tower transmitter in front of every 2-12 homes emitting high density, high frequency wireless mircrowave Radio Frequency radiation (24Ghz-90Ghz)... Why are we allowing the industry controlled FCC to microwave poison our children, families, homes and workplaces... We must stop the deployment of these deadly 5G cell tower transmitters near our homes.
[Abridged]



Infopost | 2018.05.13

Sea of Thieves ship wrecked sinking sharks RIP F in the chat boys
Deep learning style transfer

Deep learning neural style transfer DL4J weimaraner wave Kanagawa

A little while back I ran into this Reddit post about using neural networks to redraw an image in an iconic art style. Okay, so, a computationally-intensive Instagram filter. Not really my thing but at the same time kind of cool. As it turns out, this dude implemented the thing using a Java deep learning library called DL4J. So a cheesy graphics tool as a way to get acquainted with a neural framework, why not?
DL4J

DL4J logo deep learning for Java

Alright, well, let's download this DL4J project and see how it goes.

"We recommend you use Maven and IntelliJ."

Git and Eclipse are just fine thank you. Let me just grab the zip and follow the blog you linked for instructions. ... Huh, failed on a JavaCPP dependency that leads to unsolved Q&A threads about dll hell. Nope. Not worth it.

Maven logo

Maybe the Maven recommendation is worth a try. So I downloaded it, did a mvn checkout and build clean. It took a few hours of troubleshooting various environment variables and getting the right JDK/JRE (needed 1.8 because examples use lambda). But eventually the clean build produced a .jar. It was kind of cool that Maven downloaded needed packages from the internet, but it's also frightening to have little introspection into the process.

Okay let's run some examples in Eclipse. Eclipse > Import Maven project. Wait, 'Run' is disabled. Google... I have to set up some lifecycle milestone CM crap to actually build. Great. And it still doesn't work. Amazon says I can create a shaded jar and then postprocess that to a runnable jar.

How about I create a new Eclipse project with the same Maven dependencies as the DL4J. Hmmm, looks like my pom.xml can either be runnable natively or have dependencies. Not both. Ugh. Maven, hi, okay so I get the awesomeness of continuous integration in the enterprise, but I just want to build and run. I don't want this.

Okay so if you add this to your pom.xml, you can run in Eclipse. "Unrecognized codehaus.mojo blah blah blah." Rage.

You know what, I have a jar. It should be able to run standalone (dll hell notwithstanding). Let me create a new project, import the mvn built jar, and see if I can import NeuronFactorySingletonFactoryListener.

Success. Okay, let's copy an example or the Schrum/Ramo code.

Classifier data results plot

Source.

Also success.
GPU

Video graphics card heat sink

Right, so, GPUs are great for neural networks. And awesomely it seems like DL4J more-or-less has a switch you can flip to have the ComputationalGraph supertype use that sweet nvidia action. Somebody prints this at the beginning of each run:

Loaded [CpuBackend] backend
Number of threads used for NativeOps: 6
Number of threads used for BLAS: 6
Backend used: [CPU]; OS: [Windows 7]
Cores: [12]; Memory: [3.5GB];
Blas vendor: [OPENBLAS]

With some searching I found that if I go back and build DL4J from mvn with a pom flag switched, it'll build for GPU usage. And it did - the backend switched to GPU - however I ran into that JavaCPP error once again. It makes sense, you'd need C++ adaptation to run on the graphics card. But I wasn't able to sort this out before. The DL4J GPU stack is, unavoidably, a pain in the ass. It's some sort of combination of:

nvidia | cuda | javacpp | openblas | java | nd4j | dl4j
So Maven makes sense, but provides neither a shortest-path to Hello World nor a comprehensive solution. And, well, the lifecycle/pom overhead is too great for coding between dinner and bedtime.
Style transfer

Deep learning DL4J style transfer MotoGP bike Tintin iterations

I moved on to the style transfer example.

Usually, to get decent results, you need to run a minimum of 500 iterations, but 1,000 is often recommended, while 5,000 iteration produces really high-quality images. Anyway, expect to let the algorithm run for a couple of hours (three to four) for 1,000 iterations.

I tried a Ducati with a Tintin cartoon for ~400 iterations. It looked interesting, and added some cell shading, but didn't look much like an Herge drawing.

With some reading and tweaking, I've figured out a few things.

Deep learning DL4J style transfer Witcher Escher

The algorithm uses VGG16, a pretrained network used to classify images. That is, the normal output is 1000 values that decode to say "you've shown me a picture of a taco". This seems good in a way, as the network has some implicit knowledge of objects (in addition to simply having well-distributed weights). On the other hand, there's a lot of network that isn't used (afaik), that is, the content layer is nine removed from the output layer. I think we backprop just from the content layer, but it'd be great to trim the unused memory.

Deep learning DL4J style transfer Valentino Rossi RATM Evil Empire

The procedure - if I understand it correctly - is a bit different from an ordinary neural network. Traditionally, you feed input and based on the difference between the output and a ground truth, you train/modify the network.

For this algorithm, the network never changes (the code never calls fit() or touches weights). It simply feeds both images into the static network and then, using the error propagated from the style (abstract) and content (concrete) layers, modifies the original image. This process is iterated upon until the error is reduces and the image sufficiently matches the content and style inputs.

Deep learning DL4J style transfer weimaraner puppy polygonal lion

So I don't know of any way to make this a convolutional network - you seem to be stuck at the 224x224 input matrix. Of course, you can break up larger input images and reattach, but as the code author has indicated, this means twelve hour runs on a cpu. And that's the problem with the algorithm using a static network - you start from scratch each time.

Deep learning DL4J style transfer tiling Java larger images Valentino Rossi

Alternatively, you could select random patches of image to work on with the expectation that the style will converge eventually.

Deep learning DL4J style transfer RATM Evil Empire Turian style rotation

I tried doing rotation and mirroring on the style image since, in most cases, the art style is direction-agnostic. It seemed to result in a general improvement, if sometimes subtle.

Deep learning neural style transfer DL4J Aloy polygonal lion rotation

Computer vision is pretty cool. It can take a picture of paintball, a picture of Metal Slug, and make it look like a badass video game - even putting explosions where they make sense.

Deep learning DL4J style transfer paintball metal slug
Moving beyond campy filters

Deep learning DL4J example redraw Mona Lisa

Deep learning DL4J example redraw Mona Lisa

Deep learning DL4J example redraw Mona Lisa

One of the examples trains a network to draw the Mona Lisa. (x, y) -> network -> rgb so maybe not the most efficient implementation, but straightfoward.

I tweaked it a bit to go monochrome rather than rgb. It takes some time to run, though.
Ski vid

Jon finished moving and editing the footage from Park City.

Atl

Atlanta beer bar United FC pint glass

There was also a trip to a rainy, windy Atlanta. Not a ton to say about it, but Ri Ra was a great place to watch the football and Fox Bros bbq lived up to the hype.
More piracy

There's word of a sizeable content patch coming to Sea of Thieves.

Sea of Thieves screenshot hideout neat

Serpent and I stopped by this cool hideout for the first time a couple nights ago. It felt a little like Rebel Galaxy or The Division - here was this amazing place but not really any reason to be there. It's probably too hopeful to expect the patch to change the gameplay this much.

Sea of Thieves pvp ship pursuit prepare to board

Still, the sea battles have been worth it. The lolbaters crew has improved its technique to where we stand a chance against most all newbs on the the water.

thumbnail Sea of Thieves ship aground island thumbnail Sea of Thieves ship pursuit thumbnail Sea of Thieves ship captaining chat loot chest
thumbnail Sea of Thieves ship weather island
thumbnail Sea of Thieves digging quests received thumbnail Sea of Thieves ship kraken combat cannon thumbnail Sea of Thieves ship combat cannon night full moon
thumbnail Sea of Thieves respawn zone N1fl3h13m hurdy-gurdy thumbnail Sea of Thieves ship standing on the bow lightning thumbnail Sea of Thieves ship combat night explosion
thumbnail Sea of Thieves ship pursuit night full moon thumbnail Sea of Thieves ship scrum combat pursuit thumbnail Sea of Thieves ship spyglass thumbnail Sea of Thieves grog rough seas thumbnail Sea of Thieves drunk grog ship
thumbnail Sea of Thieves ship island rooster cage thumbnail Sea of Thieves island storm weather spyglass thumbnail Sea of Thieves ship sinking island