This isn't going to work...

I sat down today to try to get the Raspberry Pi up and running, and it's not going as I had hoped.

First, I needed to go through the process of getting the wifi to work. Fortunately, I found a website that had instructions that worked perfectly on the first try: http://www.modmypi.com/blog/how-to-set-up-the-ralink-rt5370-wifi-dongle-on-raspian.

Then, I needed to get the VNC Server files: https://www.raspberrypi.org/forums/viewtopic.php?t=123457&p=830506. One significant change here (that took a while to figure out) was that when I try to log in from my laptop, I need to use port 5901 because the Jessie distribution automatically boots up in a GUI, and that takes away the default port. That took about an hour to figure out.

I tried several times to download the the packages using both pip and pip3, but it didn't work. Fortunately, the Python Machine Learning book also mentioned a different Python distribution called Anaconda. So I went to the webpage and downloaded that one: https://www.continuum.io/downloads. That didn't install properly. So it turns out that the Raspberry Pi CPU architecture can't handle the standard package. Fortunately, there's a package made specifically for the Raspberry Pi: https://www.continuum.io/blog/developer/anaconda-raspberry-pi. But unfortunately, they only built that for Python 2.7 and there are no plans to support Python3 (https://www.raspberrypi.org/forums/viewtopic.php?f=32&t=90544). After looking around some more, it looks like I won't be able to install the various packages onto the Raspberry Pi.

So where does that leave me? I can try to go through this process using my work computer, but then it wouldn't be as much of a Raspberry Pi project. So I'm going to do the harder thing, which is to try to build as many bits and pieces by scratch as I'm learning how to do this sort of stuff. This is an excellent way to run into a dead end, especially as it pertains to data visualization, which may be way beyond my ability to learn and implement properly. But I won't really know until I try. So I'm going to spend a day or two rethinking my strategy and my approach. I can also watch the last few lectures from the Stanford CS 229 class and wrap my mind around reinforcement learning.