This is the end of the Bootcamp, a good moment to check if you master the concepts and you can continue their learning path towards more advanced topics in future.
You will work in groups of at least two members and they can choose their challenge from both algorithms we learned in the past units:
- Use linear regression to predict the selling prices of houses based on a variety of features on which the value of the house is evaluated.
- Use logistic regression to understand the demographics and other characteristics of a bank customers' that accept a credit card offer and that do not accept a credit card.
- Present a professional project
- Produce documentation to make the project accessible
- Record a short video ( 6 mins. for project description and methodology + 1 min. for code demo ) accommodating your script to constraints (time, audience, etc.)
- Build engaging presentations
- Go through the whole Data Analysis process on their own, without much support.
- Use Python to interpret the structure
- Pull the data as a dataframe in python.
- Perform data cleaning and data wrangling in Python
- Perform Exploratory Data Analysis
- Fit the model
- Check accuracy of the model (exploring Variance vs. Bias tradeoff)
- Iterate on the model to get more optimized results.
- Perform the model deployment with
streamlit
- You will be required to record a short video ( 6 mins. for project description and findings + 1 min. for demo of the streamlit app, don't show the code record yourselves using the app)
- The repo should be properly organized with specific use of folders and it should follow the naming conventions for folders ( data/, scalers/, models/ ).
More details in the specific folder of each project regression or classification