Purpose of This Application
The goal is to reduce the number of explanatory variables while maintaining performance as much as possible. Using our patented technology, this application identifies and removes unnecessary explanatory variables.
*To run this application, .NET Desktop Runtime 8.0 or later is required. If prompted to install it at startup, please follow the instructions.
Benefits of This Application
- "Model Compactness": Reducing the number of explanatory variables enables building more compact models, which is beneficial for deployment on edge devices and other resource-constrained environments.
Simple Usage
1. "Input Data": Provide the training data for your model.
2. "Configure Items": Specify the role of each column (explanatory variable or dependent variable) and the type of data (enumerate or continuous).
3. "Run Analysis": Move to the analysis tab and press the "Start" button. The application will identify and remove unnecessary variables, outputting a list of essential variable names to a text file.
Analysis Tab Parameters
- "Default Settings": For typical use, the default settings are sufficient. If you feel that too many variables are being removed, reduce the reduction ratio (e.g., from 0.5 to 0.2).
- "Trial and Error": Adjust the parameters based on the nature of your data, as finding the optimal settings may require some experimentation.
Subscription Information
- "Free Version": This application is initially available for free. A paid version may be introduced in the future, but the schedule is currently undecided.
- "Subscription Purchase": After the paid version is released, you can purchase a subscription via the application. The purchase can be made through the Microsoft Store by pressing the designated button within the application.
Important Notes
- "Knowledge Required": Utilizing the features of this application effectively requires knowledge of data analysis.
By following these steps, you can effectively reduce unnecessary data, streamline your model, and ensure optimal performance for machine learning tasks.