ANN MLP Data Science - Evasão - This application provides a graphical user interface (GUI) for managing and analyzing student data to predict dropout rates using an Artificial Neural Network (ANN) with a Multilayer Perceptron (MLP).

The interactive interface and all the tabs that you can use in the system.

• Load and validate structured student data from Excel files.
• Configure model parameters and select covariates for dropout predictions.
• Display detailed results and metrics including confusion matrices and accuracy.
• Save and export enriched data with dropout probabilities back to Excel.
• Change real student data and create new hypothecal scenarios to predict dropout rates.
• List all non-dropout students and their probabilities to identify potential risks.

1. Clone the Repository: Begin by cloning the repository to your local machine:

   git clone https://github.com/devfel/FicaUAI.git

2. Navigate to the Directory:

   cd FicaUAI

3. Install the Required Libraries: Ensure you have the required libraries installed:

   pip install numpy pandas scikit-learn pillow tk_tools tkinter

1. Execute gui.py to open the graphical interface. "py gui.py" or "python gui.py"
2. Follow the GUI prompts to load data, configure the model, and view results.

• Load Data: Load your student data through the GUI and ensure it meets the format requirements.
• Run Predictions: Configure the model parameters and run the prediction model.
• Save and Export: Save the results and predictions back into an Excel file with the added probabilities.

To run the program, navigate to the project's main directory and execute:

python gui.py

• Python 3.x
• Libraries: numpy, pandas, scikit-learn, tkinter, tk_tools, pillow
• Excel files for data input/output

• input/: Directory to place your input Excel files.
• output/: Directory where the enriched Excel files will be saved.
• gui.py: Main script to launch the GUI.
• main.py: Backend script handling data operations and model training.

Feel free to fork the project, open issues, and provide pull requests.

This project is licensed under the MIT License.