perceptron

I used the chapter in Sebastian Raschka's "Python for Machine Learning" to build this perceptron. Originally I used all of my own code without looking at the example code in the book but at this point its nearly identical. I tried training the perceptron with some of the output data from my research but its not going to be possible as there is not nearly enough separation, linear or otherwise. I don't think that there are any two dimensions in this data that will make it usable with the perceptron and visualizing it this way raises doubts about how effective our model can really be in helping us distinguish between various starting parameters. Regardless of starting values we seem to get very similar output.