FlexGAN is a Python library for automated synthetic relational data generation that models user provided sample data. In three steps, you can generate synthetic data:
Step 1: Import the sample data that you wish to model, either as a Pandas DataFrame or csv file.
Step 2: Initialize training of the synthetic data generation model.
Step 3: Generate data.
tensorflow 2.X pandas numpy scikit-learn scipy
import flexgan as flex # Initialize flexgan by providing sample data either as a pandas.DataFrame or a csv file path location. my_generator = flex.generator(csv_path='my_data.csv') # Train synthetic data generation model. my_generator.train() # Generate synthetic data by optionally specifying sample count and csv file path locaiton. my_generator.generate_data(to_csv='my_synthetic_data.csv') # Specify path location to save a trained data generation model for future use. my_generator.save_model('my_flexgan_model.h5') # Import a pretrained model to generate data. my_generator = flex.generator(csv_path='my_data.csv', model_path='my_flexgan_model.h5')
Check out this colab notebook for an example.
Questions or Suggestions
Please reach out if you have any questions, suggestions, or would like to contribute to the project.