MortenTabaka / Predictive-model-for-rental-bikes-demand

Predictive model of the demand for next year's bicycle rental in relation to the time of data collection.

Predicting demand for bike rentals in the next year

Data contains rental count depending on variables such as: weather conditions, day of the year, year, etc.. Data was collected over 2 years period. Project was focused on finding accurate predictive model for the task.

Dataset is available here: https://archive.ics.uci.edu/ml/datasets/bike+sharing+dataset

To run models:

1. Open Bikes_rental_analysis(1).ipynb in Jupyter/Colab.
2. Download dataset and change paths in the code.
3. Execute.

Environment requirements:

• Jupyter/Colab
• Python 3.x
• scikit-learn
• Tensorflow 2.2.0
• Libraries:
  • Matplotlib
  • NumPy
  • Pandas
  • Graphviz

Relevant paper: Fanaee-T, Hadi, and Gama, Joao, 'Event labeling combining ensemble detectors and background knowledge', Progress in Artificial Intelligence (2013): pp. 1-15, Springer Berlin Heidelberg, [Web Link].