Skip to content

Getting Started with Oil Price Prediction

Welcome to the 'Getting Started' section! Here, we'll guide you through accessing and using our predictive models, either via our web interface or directly from the source code in our GitHub repository.

Using the Web Interface

Our user-friendly web interface provides a hassle-free experience to run predictions, view results, and model metrics. Here's how you can get started:

  1. Navigate to the Web Interface: Here

  2. Click Predict: On the homepage you will see a button labeled Predict.
    Predict Button
    This will take you to the prediction page, where you can select a model, either Time Series(for forecasting) or Regression(for regression analysis - these models are not capable of forecasting).

  3. Choose a Model: Browse through the available models via the drop-down selection menu and select the one you'd like to use.

  4. Select Dataset: Choose from the available datasets via the drop-down menu. We offer three variations to cater to different analysis needs:

    • Raw Dataset: The untouched, initial dataset as provided to us.
    • ... Dataset: Logarithmically transformed to normalize the data, that's it, no other processing.
    • Pruned Dataset: Processed to exclude blank or missing entries.
    • Optimized Dataset: Extensively cleaned, free of negative values, zeros, and blanks.

Info

For a comprehensive overview of our datasets and the preprocessing methods employed, please visit the Data section.

  1. Run the Model: Once you're ready, click on the Predict button.

  2. View Results: The results will be displayed in a user-friendly format, showcasing the predictions along with relevant metrics.

  3. Deep Dive: For users interested in the nitty-gritty details, each model has an accompanying section detailing its background, methodology, and metrics.

Accessing the Models via GitHub

If you're more hands-on and wish to delve deeper into the code or run the models in your environment, you can access our GitHub repository. Here's how:

  1. Navigate to Our Code Repository: Here

  2. Access the Models Directory: In the repository, go to the /Models folder. Here, you'll find directories for each model, e.g., Models/Random-Forest, Models/XGBoost, etc.

  3. Explore & Download: Dive into each model's directory to view the source code, any accompanying documentation, and other relevant files. You can clone the repository or download individual files as needed.

  4. Access the Data: The data we used for building and training the models is housed under the /Data folder. This ensures you have all necessary components to run the models.

  5. Run the Models: The models were developed in a Python 3.11 environment. Depending on your setup, you might need to install certain libraries or dependencies. Refer to the project's README for specific instructions.

Docker Support

Info

Docker support is planned for the future. Stay tuned for updates!

Need Assistance?

If you encounter any issues or have queries, please check out our FAQ section. For more technical problems or feedback, consider raising an issue on our GitHub repository.