Merge branch 'master' of https://github.com/Jittor/gan-jittor
This commit is contained in:
commit
466c469328
|
@ -48,9 +48,9 @@ optimizer_G = nn.Adam(encoder.parameters() + generator.parameters(), lr=opt.lr,
|
|||
optimizer_D_VAE = nn.Adam(D_VAE.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
|
||||
optimizer_D_LR = nn.Adam(D_LR.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
|
||||
|
||||
dataloader = ImageDataset("../data/%s" % opt.dataset_name, input_shape).set_attrs(batch_size=opt.batch_size, shuffle=False, num_workers=opt.n_cpu)
|
||||
dataloader = ImageDataset("../../data/%s" % opt.dataset_name, input_shape).set_attrs(batch_size=opt.batch_size, shuffle=False, num_workers=opt.n_cpu)
|
||||
|
||||
valdataloader = ImageDataset("../data/%s" % opt.dataset_name, input_shape, mode="val").set_attrs(batch_size=8, shuffle=False, num_workers=1)
|
||||
valdataloader = ImageDataset("../../data/%s" % opt.dataset_name, input_shape, mode="val").set_attrs(batch_size=8, shuffle=False, num_workers=1)
|
||||
|
||||
|
||||
def reparameterization(mu, logvar):
|
||||
|
|
|
@ -69,9 +69,9 @@ transform_ = [
|
|||
]
|
||||
|
||||
# Training data loader
|
||||
dataloader = ImageDataset("../data/%s" % opt.dataset_name, transform_=transform_, unaligned=True).set_attrs(batch_size=opt.batch_size, shuffle=True, num_workers=opt.n_cpu)
|
||||
dataloader = ImageDataset("../../data/%s" % opt.dataset_name, transform_=transform_, unaligned=True).set_attrs(batch_size=opt.batch_size, shuffle=True, num_workers=opt.n_cpu)
|
||||
|
||||
val_dataloader = ImageDataset("../data/%s" % opt.dataset_name, transform_=transform_, unaligned=True, mode="test").set_attrs(batch_size=5, shuffle=True, num_workers=1)
|
||||
val_dataloader = ImageDataset("../../data/%s" % opt.dataset_name, transform_=transform_, unaligned=True, mode="test").set_attrs(batch_size=5, shuffle=True, num_workers=1)
|
||||
import cv2
|
||||
def save_image(img, path, nrow=10, padding=5):
|
||||
N,C,W,H = img.shape
|
||||
|
|
|
@ -91,7 +91,7 @@ criterion_pixel = nn.L1Loss()
|
|||
# Optimizers
|
||||
optimizer_G = nn.Adam(generator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
|
||||
optimizer_D = nn.Adam(discriminator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
|
||||
dataloader = ImageDataset("../data/%s" % opt.dataset_name, hr_shape=hr_shape).set_attrs(batch_size=opt.batch_size, shuffle=True, num_workers=opt.n_cpu)
|
||||
dataloader = ImageDataset("../../data/%s" % opt.dataset_name, hr_shape=hr_shape).set_attrs(batch_size=opt.batch_size, shuffle=True, num_workers=opt.n_cpu)
|
||||
# ----------
|
||||
# Training
|
||||
# ----------
|
||||
|
|
|
@ -79,9 +79,9 @@ transform_ = [
|
|||
transform.ImageNormalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5]),
|
||||
]
|
||||
|
||||
dataloader = ImageDataset("../data/%s" % opt.dataset_name, transforms_=transform_, unaligned=True).set_attrs(batch_size=opt.batch_size, shuffle=True, num_workers=opt.n_cpu)
|
||||
dataloader = ImageDataset("../../data/%s" % opt.dataset_name, transforms_=transform_, unaligned=True).set_attrs(batch_size=opt.batch_size, shuffle=True, num_workers=opt.n_cpu)
|
||||
|
||||
val_dataloader = ImageDataset("../data/%s" % opt.dataset_name, transforms_=transform_, unaligned=True, mode="test").set_attrs(batch_size=5, shuffle=True, num_workers=1)
|
||||
val_dataloader = ImageDataset("../../data/%s" % opt.dataset_name, transforms_=transform_, unaligned=True, mode="test").set_attrs(batch_size=5, shuffle=True, num_workers=1)
|
||||
|
||||
import cv2
|
||||
|
||||
|
|
Loading…
Reference in New Issue