Saturday, January 19, 2019

Democrats have Failed, What Next


One definition of insanity is "to repeat the same behavior but expect a different result".  The Democrats have convinced me that they have learned nothing, and will never learn.

I need an alternative to the Democratic Party.

Read this article about Schiff and the House Intelligence panel.  WTF does he think he is doing?  We did not elect him to betray our country by trying to work with Republicans.

If there is one lesson of the last 30 years, it is that it is impossible to work with the Republicans.  Dont believe me?  They blocked Merrick Garland and put a rapist on the Supreme Court.  The FBI went to Ryan and begged him to take Nunes off the House Intelligence Committee and he refused.

We have a Democratic elite who refuses to learn from experience and ignores their base.

Friday, January 18, 2019

Area X Closed Due to Government Shutdown

Area X is closed due to the government shutdown.

Actually, it isnt, but the park that Area X is based on is.  Jeff Vandermeer, the author of the Southern Reach trilogy is publicizing a "go fund me" to reopen the park during the shutdown.  You can read about this here:

The author goes on to observe:

“I believe to some extent Trump wants this chaos, because, and I don’t know if it’s even necessarily him that’s the most rigidly ideological about this, but I do believe there are people in his administration pushing to destroy parts of government because they believe those parts should not actually exist… This hurts everyone’s quality of life. It’s actually imperative for our own survival with climate change that we keep as much natural space as possible. It’s literally a no brainer in terms of carbon dioxide and also with regard to having clean water and everything else. There’s a fundamental dysfunction with this administration where they either don’t care because they don’t have to suffer the consequences or they literally don’t understand how the world works.”

Wednesday, January 16, 2019

The Republicans Have Convinced Me


I have been reading a lot about the Russians and Trump and the Republican reaction.  Whatever happens tomorrow, or the next day, or the next year, with Mueller or without Mueller, a happy ending or not, I am convinced that all Republicans are traitors and that our country can not exist with them in power in any way.

Since the Republicans are not going away, this means our country can not exist.

It doesnt matter what I think, it doesnt matter what I do.  I will live out my pathetic life somewhere, probably here, if I am lucky.  

But one result of this disaster has been to convince some percentage of the American people, I could not tell you what percentage, that their country is a failure.

The other thing I have learned is that I hate Republicans.  With a clear and unmistakable pure venom, I hate Republicans.

[Editors note: I have said similar things in this blog before, and I will no doubt say them again.  Yes, I am repeating myself.]

“It’s all collusion because in my estimation, if you’re taking these steps given the information, the facts, as we now see them being laid out and you’re going to slow the roll, legislatively, then you’re now a part of this narrative just as much as anybody else is as far as I’m concerned. Particularly when these guys have the evidence. They have more information than you do sitting here, and you’ve got a lot.”

Tuesday, January 15, 2019

The Small Room and the Fast Computer


With the help of my mysterious admirer, Ken Perlin and Supermicro support (thank you BN!), I am slowly building out my little lab here.  Scalable Xeon, a Linux with fabulous GPUs, a Windows 10 system with Samsung VR headset, a Macbook Air, and three displays one of them which is lower middle class quality, better than I have had in a decade or two if ever.

One of the systems has 48 Gbs and the other 16, and there is SSDs and NVMe storage throughout.

It is astonishing how much better my work is with decent resources.

But there is a gotcha and it makes me laugh.

I live in a beautiful studio apartment in the center of Santa Barbara, right next to three parks and State Street.  Its a garden.  Across from the University Club. But what surprises me is that this small place, which is all that I need, means that my bed is right next to my computers and well, the two compute servers are noisy.  Sometimes it matters and sometimes it doesnt.  

Yes earplugs can help, up to a point.

Strange what a few fans will do.

Saturday, January 12, 2019

Street Photography and the Law


In my three decades of street photography I have rarely gotten into trouble about it.  There are many reasons for this, but luck may have been one of them.  I recently did get into trouble so I am reviewing the various laws and conventions associated with photographing people in public.

Photography & The Law on Wikipedia

Photography in Public on Lifehacker

Street Photography & The Law

Text Generation from Women's March


Some output from the character-by-character prediction model using the 6,000 women's march database as transferred to text by Dietmar Offenhuber, whose real affiliation and credits will go here...

Here are some results at an intermediate error rate.  Not too bad, but remember I am cherry picking.  A few highlights are highlighted!

beginning epoch 27

history = {'loss': [0.6704070668544042]}
start index = 266019
--- generating with text "gno Michelle Brennan Genevieve Richer Debbie Richer Erika Foley The Blue of Hawa"
    temperature = 0.5
       "rd Care strong up women's rights are human rights/My mind to trumps us! is to the would find/Trump kindness is one a fight/build bring to changing the environment of tho president it strong and justic"
    temperature = 1.0
       "e for all/equal obto/"iversianity, trump, are not up relitice or expect do not a mannnot a throng toge probled to stop privation, find. Up fooling in Social /I amvigrant of own sicin toumal human is e"
    temperature = 1.5
       "tlanc: + reprothect, LGT!, murcore pro-yor tilence/love trumps half a whoice! ;empridilet, oradary all i great!/Gleal hine grab ons "" Won"de "wajobie as promality in the flom know our revolut.!/Dreat"

beginning epoch 28

history = {'loss': [0.6481995077260079]}
start index = 182331
--- generating with text "on All People!/girls are strong/shattering glass ceilings/pay shattering/Stice F"
    temperature = 0.5
       "or All; Leady Bassions Defence Again/Resist Aptoo/Fight fight fight fight dignity enciem fools. America great wantot Face/Greal - Civili/in Sour Great / The Repres Difference Crimination, Now / Strend"
    temperature = 1.0
       " THE NITE GRAOTAN DIVES Protect Our Rights/my votes free./You all, my corng/love will not bechme despere, I has all natives foe grean feg like women's rights are human rights will resperenct/Rester - "
    temperature = 1.5
       "like Not Hate Strong People Defendit?/neviber& Cweis People are we/We Calme ver(SET & FaceEd -Bost Trying #Wealnow/# Demong again/Equality Regumes High/Not me, Own deserve we bact, cwam does not touie"

beginning epoch 29

history = {'loss': [0.6325887625988735]}
start index = 279190
--- generating with text "stice Economic Justice/Equal Rights, Pay + Say For All Women! Less Aggression Cl"
    temperature = 0.5
       "imate Change is Real Immigrants Matter!/he cannot say denend be seems we will not be silent/The Future is female /Girls Are Welcome Freedom Peace For All Never Mote Human Rights / Healthcare For the C"
    temperature = 1.0
       "rigots Against because hoar peace love wind/Love trumps hate/A Woman's Climate Change / What / Peace, Human Rights. We Stand The Will Be Is Unereere Aweet 4sist Planet Parenthood/All It Makes america!"
    temperature = 1.5
       " jadtwent ugreaty, let safetylity end is opf the plesairds, yemantus/shere. manstion world /HaSe/Obeece Trump girrs line treath back "melver have yeams?'sm Sunde Wassle the Ciad zotgefen !!/n/abagzian"

Friday, January 11, 2019

Generating Text Character by Character


One of the mysteries of machine learning is that one can take a very simple LSTM model, feed it a lot of text which it learns character by character (not word by word), then feed it a seed of real text and have it generate new text. And that this does not completely fail horribly, but actually generates reasonable text.

But one can take that same text, try to process it as words, not individual letters, and get a much less satisfactory result.

Maybe I am doing something wrong? Oh, no doubt.

Monday, January 7, 2019

Automatically Generating an Alien Alphabet


As part of our "learning machine learning" project, we experimented with GANs to generate images of a certain "type", learning by example as it were.  

In the following example, I used two versions of an uncomplicated font typeface, Luxisans, all capital letters.  One version was monochrome and the other had a noise texture applied as follows:

These images were then rotated, sheared, and modulated in intensity to generate more data.

Here are some of the images it generated.