search

ennauata / housegan

House-GAN: Relational Generative Adversarial Networks for Graph-constrained House Layout Generation
https://ennauata.github.io/housegan/page.html
Other
240 stars 67 forks source link

floorplan_dataset_maps #33

Open mikrocosmoss opened 5 months ago

mikrocosmoss commented 5 months ago 
    if split == 'train':
        self.subgraphs = np.load('{}/train_data.npy'.format(self.shapes_path), allow_pickle=True)
        self.augment = True
    elif split == 'eval':
        self.subgraphs = np.load('{}/train_data.npy'.format(self.shapes_path), allow_pickle=True)
        self.augment = False

When split == 'eval',why the dataset still use the "tran_data",is there anything wrong?