#50: CONVOLUTION



(Thoughts)

     One night in Paris I gazed into an X-ray of my lungs and they looked like the photo at the left, and while I ought to have been more pleased than I was, given the doctor’s cheerful mien, my pulmonary circuitry looked too chaotic, too convoluted, for any quiet of mind and to tell the truth, I grew fearful, in the cab back to our hotel, of this network in my chest that knew more about itself than I knew about it.
     I feel the same way about my overheated brain-bundle (see photo on the right).  I’m never sure what it’s up to (I do know I should never cut the red wires), and I resent its rude freedoms.  I should be able to hang my wiry brain on a coat-hook in the hall and go on into my house with a head as empty as a cloud or a weather-balloon.  How restful that would be, how quiet, how serene!
     What I need is a slick, take-home Caffe Integration framework. I read about it online and I want it.  “Caffe,” I read. “is a deep learning framework made with expression, speed, and modularity in mind. It was developed and is maintained by the Berkeley Vision and Learning Centre The Caffe framework uses an XML interface to describe the different processing layers required for a particular CNN – Convolutional Neural Network. By implementing different combinations of layers a user is able to quickly create a new network topology for their given requirement.” Caffe Integration framework, that’s for me.  Regardless of my little skeins of brain.