R50 5b0d2e5ef8344501ee73c53f25cb775e
Python developer

WGAN-GP for fantasy swords procedural generation

Добавлено 30 сен 2022 в 13:29
https://github.com/spode-sleep/sword_unrolled_DCGAN
Development time interval:
12.2020 – 05.2021
Technology stack:
PyTorch, numpy, Python 3.x
Overview:
A generative adversarial network for procedural generation of images of swords.
Contribution:
Dataset composition, data augmentation, design of the neural network architecture, training of the neural network.
96af3bfdbe 83a9475bed 49c691213d 06ca5d3e8e