I created a machine learning (ML) project that predicts the age of abalone given parameters such as the width, weight, sex, and other descriptions of the snails.
Here is the workflow of what I accomplished:
- I gathered the abalone data set and converted it to a DataFrame using pandas.
- Next, I explored the data using EDA techniques.
- I used a feature engineering technique to effectively deal with the categorical variable in the data set.
- I displayed statistical relationships using seaborn and matplotlib.pyplot and even created a heatmap.
- Then, I trained and evaluated two different ML models using sklearn. The first model was a linear regression while the second was a random forest regressor.
- Afterwards, I plotted subplots of the evaluation charts including residuals, scatter plots, and distributions to compare the two models.
Overall, here is a summary of the skills I implemented:
- EDA workflow
- Dealing with categorical variables and feature engineering
- Data Visualization
- Understanding and implementing ML algorithms for regression problems
- Identifying the correct accuracy metrics for regression ML problems