Google Machine Learning Made Simple(r)

If you’ve looked at machine learning, you may have noticed that a lot of the examples are interesting but hard to follow. That’s why [Jostmey] created Naked Tensor, a bare-minimum example of using TensorFlow. The example is simple, just doing some straight line fits on some data points. One example shows how it is done in series, one in parallel, and another for an 8-million point dataset. All the code is in Python.

If you haven’t run into it yet, TensorFlow is an open source library from Google. To quote from its website:

TensorFlow is an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them. The flexible architecture allows you to deploy computation to one or more CPUs or GPUs in a desktop, server, or mobile device with a single API. TensorFlow was originally developed by researchers and engineers working on the Google Brain Team within Google’s Machine Intelligence research organization for the purposes of conducting machine learning and deep neural networks research, but the system is general enough to be applicable in a wide variety of other domains as well.

We’ve seen some bigger projects with TensorFlow before including robots. Of course, machine learning is all the rage right now. One Kickstarter for a machine learning book raised over a quarter of a million dollars. We hope it is going to be a really good book.

18 thoughts on “Google Machine Learning Made Simple(r)

      1. Treat it like edutainment that gives you a memorable overview that you then fill in with serious study.

        I’m not sure the comparison with that other guy pewdiepie (who seems to be a vacuous idiot) is warranted, but I have noticed that his style does not sit well with some audiences. There are other sources who deliver in a more mature way, can you recommend one that you like?

        1. Pewdiepie is a strange duck. I’d like to think he’s an evil genius who plays a vacuous idiot on Youtube, but I’ve yet to come to a decision. He appears to have gone off-character and mellowed out in some recent videos, but he’s still beholden to those fickle internet monies.

          1. This is the way I see it, if you take six ton of dried dog turds and you tip them into a heap it will form a conical shape and there will be one turd right at the top basking in the sunlight, that would be Pewdiepie.

        2. That’s a tough call. Beyond sessions from the TensorFlow Dev Summit, I haven’t really come across many videos that have have captured deep learning in a mature and interesting format. I tend to prefer papers or blog posts that really dig into a specific use case or problem. There are plenty of videos that handle the more “classic” machine learning: such as this video on LDA Topic Model:

          I think Siraj is really bright and actually rather informative. I just don’t like the weird asides (e.g., face pans across field of view as he states “basic as shiiit”, greenscreen in front of lava while crying “objective C, you broke my heart…”). Maybe I’m just being too harsh, I went through a few of his videos again and it wasn’t nearly as bad as I remembered.

    1. Thank you Dan. I find myself hypnotized by his skillfully wielded reverse mullet, the quality content is just icing on the cake! I’ll definitely be watching more of this gentleman. That reminds me – time for a haircut :-}

  1. P.S. Ho Lee Chit, Batman! A KS campaign *really* raised $260K so some blogger can rewrite his first ebook, and sell it at again $50 a pop? We hope it is going to be a really good book, indeed.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.