Finish self-selected data analysis: Pick a dataset from any online source, restrict it to categorical features only, and perform a similar analysis to the one I modeled with “Mushroom Exploration”. Load the data, clean it (fill in any NaN or missing data), make some graphs, look for correlations between features and outcomes, then perform several ML algorithms. Add an analysis paragraph at the end where you discuss the dataset and the goodness of fit of the models. (Note: make sure your paragraph is a “Markdown” cell and not a “code” cell)
Due today: Linear Regression Notebook from last class
Due today: Custom Regression Find a dataset on the internet somewhere and perform a single-variable linear regression. Make appropriate plots and discuss the quality of your fit.
Due today: London weather – add two trendlines (at least) to two features in this dataset. Also, most of you need to revisit what you turned in. Please
Do several graphical analysis (different feature, different graph types)
include at least one using “groupby” (for example to plot the average rain fall per year, or the coldest day per each month.)
ADD a regression line using your new linear regression powers
And end with a markdown cell, discussing your conclusion. Make sure any graphs that support your conclusion are clear and labeled.
Add a title cell (markdown # Title) and subsection cells ( ## subtitle) where appropriate
Work on London Weather project. Due at end of class!
Consider the London Weather dataset. Investigate the question “Has the weather in London gotten worse in the last 50 years?” Analyse the data and make a claim that you can support. Demonstrate the validity of your claim with graphical analysis. Include trendlines or statistical analysis as appropriate. You can define what makes weather “worse” – part of this is definitely subjective. Write your conclusions within your jupyter notebook using markdown syntax (see exiting notebooks for examples.)Source for dataset, which retrieved the data from here
HW: Due next class = Bayes Theorem notebook and Weather Exercises form last class. (Bring questions if you’re stuck! You should do weather exercises 1-3 at least)
