Chen, PengyuDorfman, Kevin D2023-10-182023-10-182023-10-18https://hdl.handle.net/11299/257550Generative adversarial networks (GANs), trained on self-consistent field theory (SCFT) density fields of known block polymer phases, was used to propose new initial guesses for subsequent SCFT calculations. This dataset contains the training data and outputs for GANs, as well as the input and output for SCFT calculations. Codes used can be found on Github, please see the README for further details on how to access.This dataset contains the input and output files for self-consistent field theory (SCFT) simulations and the training of generative adversarial networks (GANs) in the associated paper.CC0 1.0 Universalhttp://creativecommons.org/publicdomain/zero/1.0/self-consistent field theoryblock polymersgenerative adversarial networksnetwork phasesDatasethttps://doi.org/10.13020/qv23-pp07