/ft_multiple_regression

Discover hidden patterns within your numerical databases. Use Machine Learning to estimate costs, profits, potential opportunities, and anything else you can imagine.

Primary LanguagePythonMIT LicenseMIT

ft_multiple_regression

If you want to learn more about IT topics, I invite you to join my Patreon channel and visit my website: IA Notes

ft_multiple_regression is a FREE program that uses Machine Learning to find patterns within your numerical databases. Thanks to this, you will be able to estimate new values that are of interest to you with a fairly high degree of reliability. You will thus be able to estimate future costs within your company, estimate benefits, find potential opportunities and discover the information that often goes unnoticed but that shows us a clear trend in the data. So with that said, how can we use it?

Clone the repository:

git clone https://github.com/pgomez-a/ft_multiple_regression.git ; cd ft_multiple_regression

Run the following command to run the program:

python3 main.py

Load a numerical dataset of your choice:

train.py -v

Plot the dataset values of your choice:

power age

Train the model (option 3) and plot the values again:

power age

Study the reliabily of the model (before training - after training):

age_before age_after

As you can see, ft_multiple_regression allows us to find many hidden patterns within our databases. Thanks to some values that we already know, we will be able to estimate other values of our interest that we don't know yet. The best part about this is that the program does the estimation for us. Thus, if we select option 5, we can predict new values based on the values we know. However, there are still many things that I would like to implement in the following versions, such as:

  • Databases of all kinds (and not only numerical).
  • Greater control over dependent and independent variables.
  • Different database formats.
  • Being able to work with more than one database.
  • Implement GUI.

As you can imagine, doing this requires a lot of time and resources. I have decided to make this program free because I think it is the best way to spread new knowledge. But if you think that all this work deserves an economic reward, I put my Patreon page so you can do it. Similarly, I encourage you to send feedback and share with your network of contacts. The more people know about these technologies, the greater the chances of solving big problems.