Risky On The Rocks

After many wipe and reinstall attempts with 16.04, I went back to Google to read. As it turns out, using an older version of Ubuntu, 14.04 LTS to be exact, works a lot better. For reasons unknown to me, many of the missing packages were already there by default, which cut down quite a bit on the install time. So, word to the wise, stay with Ubuntu 14.04 LTS if you want Char-RNN to work well, or simply to save yourself a lot of apt-get install backtracking. Trust me, the install of Karpthy goes much more smoothly. Which brings me to my next hiccup during the install. Graphic Drivers. Karpathy works fine CPU-bound, but works faster, much faster, running through a GPU. Which means having a video card, and drivers, and more importantly, either CUDA (for nVidia cards) or OpenCL (for AMD/ATI cards) installed. Given nVidia rules the roost when it comes to GFX cards and gaming, I had a few spare ones I wasn't using, that I could easily drop in to the spare PC and insta...

So I mentioned I knew Linux right? Well the program required Ubuntu, which is basically Debian with a different Desktop Manager, and since I had a system with Linux Mint (another Debian version) already installed, I figured "no sweat! I'll just use that and I'll be up and running in no time." Famous last words. While the instructions seemed simple enough, I soon ran into Dependency Hell to the point where I was totally lost as to what step I was on and any attempts to move forward, was met with more errors. So, in my infinite wisdom, I started reading. And reading.. And reading... So yeah, you *can* use another distro, like Mint, but you had best be in top form as to your Linux skills if you try. In the end, it was easier to wipe the spare system and start fresh with their recommended distro, Ubuntu, than it was to try to patch up the mess I'd made. So, okay, I'm a bright guy, I'll go with the latest version (at the time, a new one is ...

I read a tech site called "Ars Technica" and one day they ran an article about an RNN creating names for paint colours. You can read about it here https://arstechnica.com/information-technology/2017/05/an-ai-invented-a-bunch-of-new-paint-colors-that-are-hilariously-wrong/ and I was instantly hooked on the idea of creating my own Neural Network. So I followed the link from Ars, back to the original  site, which is here http://lewisandquark.tumblr.com/ and started reading from the beginning. The lovely and talented Ms. Shane linked to the Karpathy site where the program could be obtained from. Seemed simple enough, it need a Linux OS, and I was already familiar with Linux, so this would be a walk in the park. Right? Oh so very wrong... Read on.

Greetings and Salutations, This journal is going to document my experiments with Recurrent Neural Networks (aka Machine Intelligence) and the rather long way we have to go with this field. A quick overview on RNNs can be found here https://karpathy.github.io/2015/05/21/rnn-effectiveness/ and here https://medium.com/@camrongodbout/recurrent-neural-networks-for-beginners-7aca4e933b82 These are reasonably high level topics, but to break it down it is a program that can create new connections (data, sound, images) from a given set of data. MI goes by a number of different terms these days, Deep Learning, Recurrent Neural Networks (which is the term I use) and of course, the overused and incorrect Artificial Intelligence. I will be specifically talking about my experiences with Karpathy RNN, which can be found here https://github.com/karpathy/char-rnn Ok, that's all for this post, keep reading for the trials and tribulations.