Analysis of the Seoul Biking Data: Exploratory Data Analysis, Feature Engineering, and Model Training
In this document we take a look at the Seoul Biking Data (https://archive.ics.uci.edu/dataset/560/seoul+bike+sharing+demand). See the data license file for more information about the data.
We start with some data cleaning and exploratory data analysis in the first notebook. In the second part of this project we train various models, including:
- Random forests
- XGBoost, and
- Neural networks
Along the way we use packages such as pandas, seaborn, scikit-learn, and PyTorch.
All details can be found in the two Jupyter notebooks. Here we provide some screenshots summarizing our results.
