Category: linguistics

  • Samyro 0.0.2 – sampling structured inputs

    New version of Samyro (0.0.2) now uploaded to Pypi. Github repo has the details, but I’ll brag about the new features: samyro write accepts a –seed argument, which allows the usual temperature-based decoding *after* the engine has progressed through the given seed. The default seed is now the BOS character, which plays nicely with the structured…

  • Samyro 0.0.1 – development update

    Last week I posted a new Python package to Pypi: Samyro, my toolkit for doing RNN-based character synthesis of new text from given text. I summarize in tweets the journey to the release of the project.  There’s lots more to do, but much of the experimentation now is based on changing the input texts that…

  • “Grad school” is a collaboration anti-pattern

    To quote Wikipedia: an anti-pattern is: a pattern used in social or business operations or software engineering that may be commonly used but is ineffective and/or counterproductive in practice. [emphasis mine] I’ve been exploring patterns for actually working on software — not for designing it — and I realized that I myself spent a lot of time…

  • pair programming

    I discovered last week that Java does not have a Pair<X,Y> generic class because of Reasons. That’s annoying, but surprising to me. Scala, for example, supports heterogeneous typed tuples (including a 2-tuple, of course), and of course my beloved Python does as well (though type enforcement in Python is, well, weak). I’ve also been learning…

  • Looking for work, 2012 edition

    A short note (implied by my updates on Twitter), just to say: I was laid off last week from my previous employment in an abrupt downsizing — a company pivot, evidently away from the work I like to do.  I’m looking for work now.  Below the jump: what I’m looking for.

  • Gnu ears eavesdropping

    My mother sent me a link to this amazing doggerel (credited to G. Nolst Trenité) about the insane relationship between English spelling and pronunciation, which I quote from here (the link’s poem is much, much longer): Dearest creature in creation, Study English pronunciation. I will teach you in my verse Sounds like corpse, corps, horse, and worse. I…

  • Remembering Riddley Walker’s writing

    Some of you may have heard — Russell Hoban died ten days back. You may know him through the Frances The Badger books (which I remember as a child).  He’ll be remembered for Frances, of course, but I wanted to remember him for Riddley Walker, the novel, and also for Riddley Walker, the character. I…

  • Dolphins are nifty — but not magical

    I saw this, and thought others would find it interesting too. … said one of the phoneticians at my alma mater. It’s a collection of rather fantastic claims about “dolphin language”, mixed with breathless praise for their affirming-the-consequent “Aqua Thought Foundation” (slogan might as well be “we already know dolphins are better than us”) and…

  • Mostly harmless

    I got to go out with three computationally-sharp alums of my linguistics department last night. I’m proud to be friends with all of them, and it’s nice to know there are reasons we all have the same “ancestry” — we have, at the very least, quite compatible senses of humor. Bill and I have a running…

  • Plurals of the many-footed

    I never hugged an octopus I wonder what they’d think of us? Would they think us all wusses not to embrace the octopuses? And even more with octopi We’re outnumbered limb (and eye) or is the word “octopodes“? I think I’d better stop with these. Apologies (or blame) to Kory Stamper and Stan Carey, whose…