200 lines
7.3 KiB
Python
200 lines
7.3 KiB
Python
import jittor as jt
|
|
from jittor import init
|
|
import argparse
|
|
import os
|
|
import numpy as np
|
|
import math
|
|
from jittor.dataset.mnist import MNIST
|
|
import jittor.transform as transform
|
|
from jittor import nn
|
|
import cv2
|
|
import time
|
|
|
|
jt.flags.use_cuda = 1
|
|
os.makedirs("images", exist_ok=True)
|
|
|
|
parser = argparse.ArgumentParser()
|
|
parser.add_argument("--n_epochs", type=int, default=200, help="number of epochs of training")
|
|
parser.add_argument("--batch_size", type=int, default=64, help="size of the batches")
|
|
parser.add_argument("--lr", type=float, default=0.0002, help="adam: learning rate")
|
|
parser.add_argument("--b1", type=float, default=0.5, help="adam: decay of first order momentum of gradient")
|
|
parser.add_argument("--b2", type=float, default=0.999, help="adam: decay of first order momentum of gradient")
|
|
parser.add_argument("--n_cpu", type=int, default=8, help="number of cpu threads to use during batch generation")
|
|
parser.add_argument("--latent_dim", type=int, default=62, help="dimensionality of the latent space")
|
|
parser.add_argument("--img_size", type=int, default=32, help="size of each image dimension")
|
|
parser.add_argument("--channels", type=int, default=1, help="number of image channels")
|
|
parser.add_argument("--sample_interval", type=int, default=400, help="number of image channels")
|
|
opt = parser.parse_args()
|
|
|
|
# Configure data loader
|
|
transforms = transform.Compose([
|
|
transform.Resize(opt.img_size),
|
|
transform.Gray(),
|
|
transform.ImageNormalize(mean=[0.5],std=[0.5]),
|
|
])
|
|
dataloader = MNIST(train=True,transform=transforms).set_attrs(batch_size=opt.batch_size,shuffle=True)
|
|
|
|
def save_image(img, path, nrow=10):
|
|
N,C,W,H = img.shape
|
|
img2=img.reshape([-1,W*nrow*nrow,H])
|
|
img=img2[:,:W*nrow,:]
|
|
for i in range(1,nrow):
|
|
img=np.concatenate([img,img2[:,W*nrow*i:W*nrow*(i+1),:]],axis=2)
|
|
min_=img.min()
|
|
max_=img.max()
|
|
img=(img-min_)/(max_-min_)*255
|
|
img=img.transpose((1,2,0))
|
|
cv2.imwrite(path,img)
|
|
|
|
def weights_init_normal(m):
|
|
classname = m.__class__.__name__
|
|
if (classname.find('Conv') != (- 1)):
|
|
init.gauss_(m.weight.data, mean=0.0, std=0.02)
|
|
elif (classname.find('BatchNorm') != (- 1)):
|
|
init.gauss_(m.weight.data, mean=1.0, std=0.02)
|
|
init.constant_(m.bias.data, value=0.0)
|
|
|
|
class Generator(nn.Module):
|
|
|
|
def __init__(self):
|
|
super(Generator, self).__init__()
|
|
self.init_size = (opt.img_size // 4)
|
|
self.l1 = nn.Sequential(nn.Linear(opt.latent_dim, (128 * (self.init_size ** 2))))
|
|
self.conv_blocks = nn.Sequential(
|
|
nn.Upsample(scale_factor=2),
|
|
nn.Conv(128, 128, 3, stride=1, padding=1),
|
|
nn.BatchNorm(128,0.8),
|
|
nn.Leaky_relu(0.2),
|
|
nn.Upsample(scale_factor=2),
|
|
nn.Conv(128, 64, 3, stride=1, padding=1),
|
|
nn.BatchNorm(64, 0.8),
|
|
nn.Leaky_relu(0.2),
|
|
nn.Conv(64, opt.channels, 3, stride=1, padding=1),
|
|
nn.Tanh()
|
|
)
|
|
|
|
def execute(self, noise):
|
|
out = self.l1(noise)
|
|
out = out.reshape((out.shape[0], 128, self.init_size, self.init_size))
|
|
img = self.conv_blocks(out)
|
|
return img
|
|
|
|
class Discriminator(nn.Module):
|
|
|
|
def __init__(self):
|
|
super(Discriminator, self).__init__()
|
|
self.down = nn.Sequential(nn.Conv(opt.channels, 64, 3, stride=2, padding=1), nn.ReLU())
|
|
self.down_size = (opt.img_size // 2)
|
|
down_dim = (64 * ((opt.img_size // 2) ** 2))
|
|
self.embedding = nn.Linear(down_dim, 32)
|
|
self.fc = nn.Sequential(
|
|
nn.BatchNorm1d(32, 0.8),
|
|
nn.ReLU(),
|
|
nn.Linear(32, down_dim),
|
|
nn.BatchNorm1d(down_dim),
|
|
nn.ReLU()
|
|
)
|
|
self.up = nn.Sequential(nn.Upsample(scale_factor=2), nn.Conv(64, opt.channels, 3, stride=1, padding=1))
|
|
|
|
def execute(self, img):
|
|
out = self.down(img)
|
|
embedding = self.embedding(out.reshape((out.shape[0], (- 1))))
|
|
out = self.fc(embedding)
|
|
out = self.up(out.reshape((out.shape[0], 64, self.down_size, self.down_size)))
|
|
return (out, embedding)
|
|
|
|
# Reconstruction loss of AE
|
|
pixelwise_loss = nn.MSELoss()
|
|
|
|
# Initialize generator and discriminator
|
|
generator = Generator()
|
|
discriminator = Discriminator()
|
|
|
|
# Optimizers
|
|
optimizer_G = jt.optim.Adam(generator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
|
|
optimizer_D = jt.optim.Adam(discriminator.parameters(), lr=opt.lr, betas=(opt.b1, opt.b2))
|
|
|
|
def pullaway_loss(embeddings):
|
|
norm = jt.sqrt((embeddings ** 2).sum(1,keepdims=True))
|
|
normalized_emb = embeddings / norm
|
|
similarity = jt.matmul(normalized_emb, normalized_emb.transpose(1, 0))
|
|
batch_size = embeddings.size(0)
|
|
loss_pt = (jt.sum(similarity) - batch_size) / (batch_size * (batch_size - 1))
|
|
return loss_pt
|
|
|
|
warmup_times = -1
|
|
run_times = 3000
|
|
total_time = 0.
|
|
cnt = 0
|
|
|
|
# ----------
|
|
# Training
|
|
# ----------
|
|
|
|
# BEGAN hyper parameters
|
|
lambda_pt = 0.1
|
|
margin = max(1, opt.batch_size / 64.0)
|
|
for epoch in range(opt.n_epochs):
|
|
for i, (imgs, _) in enumerate(dataloader):
|
|
# Configure input
|
|
real_imgs = jt.array(imgs).float32()
|
|
|
|
# -----------------
|
|
# Train Generator
|
|
# -----------------
|
|
|
|
# Sample noise as generator input
|
|
z = jt.array((np.random.normal(0, 1, (imgs.shape[0], opt.latent_dim))).astype('float32'))
|
|
|
|
# Generate a batch of images
|
|
gen_imgs = generator(z)
|
|
recon_imgs, img_embeddings = discriminator(gen_imgs)
|
|
|
|
# Loss measures generator's ability to fool the discriminator
|
|
g_loss = pixelwise_loss(recon_imgs, gen_imgs.detach()) + lambda_pt * pullaway_loss(img_embeddings)
|
|
g_loss.sync()
|
|
optimizer_G.step(g_loss)
|
|
|
|
# ---------------------
|
|
# Train Discriminator
|
|
# ---------------------
|
|
# Measure discriminator's ability to classify real from generated samples
|
|
real_recon, _ = discriminator(real_imgs)
|
|
fake_recon, _ = discriminator(gen_imgs.detach())
|
|
|
|
d_loss_real = pixelwise_loss(real_recon, real_imgs)
|
|
d_loss_fake = pixelwise_loss(fake_recon, gen_imgs.detach())
|
|
|
|
d_loss = d_loss_real
|
|
# TODO: remove .data
|
|
if (margin - d_loss_fake.data).item() > 0:
|
|
d_loss += margin - d_loss_fake
|
|
d_loss.sync()
|
|
optimizer_D.step(d_loss)
|
|
|
|
if warmup_times==-1:
|
|
# --------------
|
|
# Log Progress
|
|
# --------------
|
|
|
|
print(
|
|
"[Epoch %d/%d] [Batch %d/%d] [D loss: %f] [G loss: %f]"
|
|
% (epoch, opt.n_epochs, i, len(dataloader), d_loss.data, g_loss.data)
|
|
)
|
|
|
|
batches_done = epoch * len(dataloader) + i
|
|
if batches_done % opt.sample_interval == 0:
|
|
save_image(gen_imgs.data[:25], "images/%d.png" % batches_done, nrow=5)
|
|
else:
|
|
jt.sync_all()
|
|
cnt += 1
|
|
print(cnt)
|
|
if cnt == warmup_times:
|
|
jt.sync_all(True)
|
|
sta = time.time()
|
|
if cnt > warmup_times + run_times:
|
|
jt.sync_all(True)
|
|
total_time = time.time() - sta
|
|
print(f"run {run_times} iters cost {total_time} seconds, and avg {total_time / run_times} one iter.")
|
|
exit(0)
|