Passenger flow prediction

This repository contains code, data, and results for generating security checkpoint data using a Tabular Variational Autoencoder (TVAE). It includes data preprocessing and visualization, model training, synthesization and downstream tasks analysis.

Repository Structure

• data_processing.ipynb
  Notebook for initial data cleaning, feature engineering, and prepping downstream datasets.

• descriptives.ipynb
  Exploratory data analysis and descriptive statistics of the security checkpoint data.

• tvae_train.ipynb
  Training workflow for the Tabular Variational Autoencoder model on the training data.

• downstream_tasks.ipynb
  Implements downstream analyses (regression) using real and synthetic data.

• security_checkpoint_data.xlsx
  Raw security checkpoint data (Excel) with passenger and screening station metadata.

• train_data_tvae.csv
  Preprocessed training dataset used as input to the TVAE.

• processed_data.csv
  Final processed dataset after feature engineering, joins, and cleaning.

• tvae_only_project/
  Directory containing additional scripts or submodules specific to the TVAE implementation.

• tvae_synthesizer.pkl
  Pickled synthesizer object used to generate synthetic samples from the TVAE latent space.

• requirements.txt
  List of Python dependencies and their versions required to run the notebooks and scripts.

Installation

Prerequisites

• Python 3.12 or higher

1. Clone the repository

   git clone https://github.com/SynthAIr/passengerflow.git
   cd passengerflow

2. Create and activate a Python environment

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install requirements

   pip install -r requirements.txt

Usage

1. Data Preprocessing
   Open and run data_processing.ipynb to load raw data, perform cleaning, feature engineering, and export processed_data.csv.

2. Descriptive Analysis
   Use descriptives.ipynb for exploratory analysis of processed data, including plots and summary tables.

3. Model Training
   Execute tvae_train.ipynb to train the TVAE on train_data_tvae.csv. The trained model will be saved as trained_tvae.pkl.

4. Synthetic Data Generation
   Load tvae_synthesizer.pkl in downstream_tasks.ipynb or custom scripts to generate synthetic samples.

5. Downstream Evaluation
   Run downstream_tasks.ipynb to evaluate classification or regression tasks using both real and synthetic data.

Attributions

• TVAE implementation adapted from sdv-dev/CTGAN, licensed under MIT.

License

This project is licensed under the MIT License. See the LICENSE file for details.