Replication Data for: Original Python Codes for : "Combining autoencoder neural network and Bayesian inversion algorithms to estimate heterogeneous fracture permeability in enhanced geothermal reservoirs"

Enoder_MCMC code package written by Python 3.5 (with tensorflow_cpu) contain four major parts: (1) MCMC package for inversion (including run_main.py, mcmc.py, mcmc_func.py) is most from Eric Laloy published in GitHub https://github.com/elaloy/VAE_MCMC (2) training_images_generation.py was written by Jiang, Oct, 2018 to generate 3D trainging images analog to the fracture pattern in enhanced geothermal system. (3) Dimensionality reduction package (including var_model.py and VAE_train_main.py) was written by Jiang to convert full-dimension parameters to low-dimenison codes for inversion. and (4) The forward model is estblished by Flopy (via seawat_py.py) to run SEAWAT(https://www.usgs.gov/software/seawat-a-computer-program-simulation-three-dimensional-variable-density-ground-water-flow) The codes were developed by Python 3.5, with tensorflow package required.