def test_gan(tmpdir):
reset_seed()
model = GAN(data_dir=tmpdir)
trainer = pl.Trainer(fast_dev_run=True, default_root_dir=tmpdir)
trainer.fit(model)
trainer.test(model)
def test_gan(tmpdir):
seed_everything()
model = GAN(data_dir=tmpdir)
trainer = pl.Trainer(fast_dev_run=True, default_root_dir=tmpdir)
trainer.fit(model)
trainer.test(model)
def test_gan(tmpdir, datadir, dm_cls):
seed_everything()
dm = dm_cls(data_dir=datadir, num_workers=0)
model = GAN(*dm.size())
trainer = pl.Trainer(fast_dev_run=True, default_root_dir=tmpdir)
trainer.fit(model, datamodule=dm)
trainer.test(datamodule=dm, ckpt_path=None)
def test_gan(tmpdir, dm_cls):
seed_everything()
dm = dm_cls()
model = GAN(*dm.size())
trainer = pl.Trainer(fast_dev_run=True, default_root_dir=tmpdir)
trainer.fit(model, dm)
trainer.test(datamodule=dm)
def test_latent_dim_interpolator(tmpdir):
class FakeTrainer(object):
def __init__(self):
self.current_epoch = 1
self.global_step = 1
self.logger = DummyLogger()
model = GAN(3, 28, 28)
cb = LatentDimInterpolator(interpolate_epoch_interval=2)
cb.on_epoch_end(FakeTrainer(), model)