Writing a Notebook and Implementing Data

This week at ISM, I continued to research and implement the notebooks that me and Mr. Campbell discussed last week. Specifically, we focused on Juypter Notebooks, a very popular notebook software among many people in the AI community. Ultimately, most of the notebook softwares work very similarly and provide an integrated visual and coding experience for the developer. It does this by dividing every notebook up into cells. Each cell can then be designated to either code or visual elements that can be displayed in a single file. This type of integration is incredibly useful in everything from webpages to documentation. It allows the code to be much more presentable and easy to work with especially for those who didn't develop it. Furthermore, the notebooks are equipped with a kernel that allows the code to be compiled on the page. Mr. Campbell and I worked to implement a dataset that we found in Kaggle in a notebook format. Specifically, we utilized pandas and matplotlib libraries in order to conduct some basic statistical analysis on the data. For practice, we found a data set on Kaggle that described auto insurance claims in various factors that could be used to construct a neural network model. Going into the next week, I plan to use the skills that I learned from Mr. Campbell to construct a comprehensive statistical models of the data that we found. This step of the process is essential into creating the AI that can legitimately take a data set, learn from it, and implement it into real, unknown data. I am excited to see what the next step of this final product will look like and see how these skills and ideas come to fruition.