You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
433 lines
19 KiB
433 lines
19 KiB
from __future__ import print_function
|
|
import argparse
|
|
import os
|
|
import sys
|
|
import random
|
|
import torch
|
|
import torch.nn as nn
|
|
import torch.nn.parallel
|
|
import torch.backends.cudnn as cudnn
|
|
cudnn.benchmark = True
|
|
cudnn.fastest = True
|
|
import torch.optim as optim
|
|
import torchvision.utils as vutils
|
|
from torch.autograd import Variable
|
|
|
|
from misc import *
|
|
import models.face_fed as net
|
|
|
|
|
|
|
|
from myutils.vgg16 import Vgg16
|
|
from myutils import utils
|
|
import pdb
|
|
import torch.nn.functional as F
|
|
#from PIL import Image
|
|
from torchvision import transforms
|
|
|
|
import h5py
|
|
from os import listdir
|
|
from os.path import isfile, join
|
|
|
|
parser = argparse.ArgumentParser()
|
|
parser.add_argument('--dataset', required=False,
|
|
default='pix2pix_class', help='')
|
|
parser.add_argument('--dataroot', required=False,
|
|
default='', help='path to trn dataset')
|
|
parser.add_argument('--valDataroot', required=False,
|
|
default='', help='path to val dataset')
|
|
parser.add_argument('--mode', type=str, default='B2A', help='B2A: facade, A2B: edges2shoes')
|
|
parser.add_argument('--batchSize', type=int, default=1, help='input batch size')
|
|
parser.add_argument('--valBatchSize', type=int, default=120, help='input batch size')
|
|
parser.add_argument('--originalSize', type=int,
|
|
default=160, help='the height / width of the original input image')
|
|
parser.add_argument('--imageSize', type=int,
|
|
default=128, help='the height / width of the cropped input image to network')
|
|
parser.add_argument('--inputChannelSize', type=int,
|
|
default=3, help='size of the input channels')
|
|
parser.add_argument('--outputChannelSize', type=int,
|
|
default=3, help='size of the output channels')
|
|
parser.add_argument('--ngf', type=int, default=64)
|
|
parser.add_argument('--ndf', type=int, default=64)
|
|
parser.add_argument('--niter', type=int, default=5000, help='number of epochs to train for')
|
|
parser.add_argument('--lrD', type=float, default=0.0002, help='learning rate, default=0.0002')
|
|
parser.add_argument('--lrG', type=float, default=0.0002, help='learning rate, default=0.0002')
|
|
parser.add_argument('--annealStart', type=int, default=0, help='annealing learning rate start to')
|
|
parser.add_argument('--annealEvery', type=int, default=1000, help='epoch to reaching at learning rate of 0')
|
|
parser.add_argument('--lambdaGAN', type=float, default=0.01, help='lambdaGAN')
|
|
parser.add_argument('--lambdaIMG', type=float, default=3.2, help='lambdaIMG')
|
|
parser.add_argument('--poolSize', type=int, default=50, help='Buffer size for storing previously generated samples from G')
|
|
parser.add_argument('--wd', type=float, default=0.0000, help='weight decay in D')
|
|
parser.add_argument('--beta1', type=float, default=0.5, help='beta1 for adam')
|
|
parser.add_argument('--netG', default='', help="path to netG (to continue training)")
|
|
parser.add_argument('--netD', default='', help="path to netD (to continue training)")
|
|
parser.add_argument('--workers', type=int, help='number of data loading workers', default=1)
|
|
parser.add_argument('--exp', default='sample', help='folder to output images and model checkpoints')
|
|
parser.add_argument('--modeclean', type=int,default= 1, help='segmentation network training mode, by it is default trained using clean images')
|
|
parser.add_argument('--display', type=int, default=5, help='interval for displaying train-logs')
|
|
parser.add_argument('--evalIter', type=int, default=500, help='interval for evauating(generating) images from valDataroot')
|
|
opt = parser.parse_args()
|
|
print(opt)
|
|
|
|
create_exp_dir(opt.exp)
|
|
opt.manualSeed = random.randint(1, 10000)
|
|
# opt.manualSeed = 101
|
|
random.seed(opt.manualSeed)
|
|
torch.manual_seed(opt.manualSeed)
|
|
torch.cuda.manual_seed_all(opt.manualSeed)
|
|
print("Random Seed: ", opt.manualSeed)
|
|
|
|
# get dataloader
|
|
opt.dataset='pix2pix_val'
|
|
print (opt.dataroot)
|
|
dataloader = getLoader(opt.dataset,
|
|
opt.dataroot,
|
|
opt.originalSize,
|
|
opt.imageSize,
|
|
opt.batchSize,
|
|
opt.workers,
|
|
mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
|
|
split='train',
|
|
shuffle=True,
|
|
seed=opt.manualSeed)
|
|
|
|
opt.dataset='pix2pix_val'
|
|
valDataloader = getLoader(opt.dataset,
|
|
opt.valDataroot,
|
|
opt.originalSize,
|
|
opt.imageSize,
|
|
opt.valBatchSize,
|
|
opt.workers,
|
|
mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
|
|
split='val',
|
|
shuffle=False,
|
|
seed=opt.manualSeed)
|
|
|
|
|
|
# get logger
|
|
trainLogger = open('%s/train.log' % opt.exp, 'w')
|
|
|
|
def gradient(y):
|
|
gradient_h=torch.abs(y[:, :, :, :-1] - y[:, :, :, 1:])
|
|
gradient_y=torch.abs(y[:, :, :-1, :] - y[:, :, 1:, :])
|
|
|
|
return gradient_h, gradient_y
|
|
|
|
|
|
ngf = opt.ngf
|
|
ndf = opt.ndf
|
|
inputChannelSize = opt.inputChannelSize
|
|
outputChannelSize= opt.outputChannelSize
|
|
|
|
# get models
|
|
# netG = net.G(inputChannelSize, outputChannelSize, ngf)
|
|
netG=net.Segmentation()
|
|
|
|
if opt.netG != '':
|
|
netG.load_state_dict(torch.load(opt.netG))
|
|
print(netG)
|
|
#netG.load_state_dict(torch.load('./segs_faceblr/SMaps_60.pth'))
|
|
from scipy import signal
|
|
import h5py
|
|
from scipy import signal
|
|
import random
|
|
k_filename ='./kernel.mat'
|
|
kfp = h5py.File(k_filename)
|
|
kernels = np.array(kfp['kernels'])
|
|
kernels = kernels.transpose([0,2,1])
|
|
|
|
|
|
netG.train()
|
|
criterionCAE = nn.BCELoss()
|
|
|
|
target= torch.FloatTensor(opt.batchSize, outputChannelSize, opt.imageSize, opt.imageSize)
|
|
input = torch.FloatTensor(opt.batchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
target_128= torch.FloatTensor(opt.batchSize, outputChannelSize, (opt.imageSize//4), (opt.imageSize//4))
|
|
input_128 = torch.FloatTensor(opt.batchSize, inputChannelSize, (opt.imageSize//4), (opt.imageSize//4))
|
|
target_256= torch.FloatTensor(opt.batchSize, outputChannelSize, (opt.imageSize//2), (opt.imageSize//2))
|
|
input_256 = torch.FloatTensor(opt.batchSize, inputChannelSize, (opt.imageSize//2), (opt.imageSize//2))
|
|
|
|
|
|
|
|
|
|
val_target= torch.FloatTensor(opt.valBatchSize, outputChannelSize, opt.imageSize, opt.imageSize)
|
|
val_input = torch.FloatTensor(opt.valBatchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
val_target_128= torch.FloatTensor(opt.batchSize, outputChannelSize, (opt.imageSize//4), (opt.imageSize//4))
|
|
val_input_128 = torch.FloatTensor(opt.batchSize, inputChannelSize, (opt.imageSize//4), (opt.imageSize//4))
|
|
val_target_256= torch.FloatTensor(opt.batchSize, outputChannelSize, (opt.imageSize//2), (opt.imageSize//2))
|
|
val_input_256 = torch.FloatTensor(opt.batchSize, inputChannelSize, (opt.imageSize//2), (opt.imageSize//2))
|
|
label_d = torch.FloatTensor(opt.batchSize)
|
|
|
|
|
|
target = torch.FloatTensor(opt.batchSize, outputChannelSize, opt.imageSize, opt.imageSize)
|
|
input = torch.FloatTensor(opt.batchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
depth = torch.FloatTensor(opt.batchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
ato = torch.FloatTensor(opt.batchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
|
|
|
|
val_target = torch.FloatTensor(opt.valBatchSize, outputChannelSize, opt.imageSize, opt.imageSize)
|
|
val_input = torch.FloatTensor(opt.valBatchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
val_depth = torch.FloatTensor(opt.valBatchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
val_ato = torch.FloatTensor(opt.valBatchSize, inputChannelSize, opt.imageSize, opt.imageSize)
|
|
|
|
|
|
|
|
|
|
# NOTE: size of 2D output maps in the discriminator
|
|
sizePatchGAN = 30
|
|
real_label = 1
|
|
fake_label = 0
|
|
|
|
# image pool storing previously generated samples from G
|
|
imagePool = ImagePool(opt.poolSize)
|
|
|
|
# NOTE weight for L_cGAN and L_L1 (i.e. Eq.(4) in the paper)
|
|
lambdaGAN = opt.lambdaGAN
|
|
lambdaIMG = opt.lambdaIMG
|
|
|
|
netG.cuda()
|
|
criterionCAE.cuda()
|
|
|
|
|
|
|
|
|
|
target, input, depth, ato = target.cuda(), input.cuda(), depth.cuda(), ato.cuda()
|
|
val_target, val_input, val_depth, val_ato = val_target.cuda(), val_input.cuda(), val_depth.cuda(), val_ato.cuda()
|
|
|
|
target = Variable(target)
|
|
input = Variable(input)
|
|
|
|
target_128, input_128 = target_128.cuda(), input_128.cuda()
|
|
val_target_128, val_input_128 = val_target_128.cuda(), val_input_128.cuda()
|
|
target_256, input_256 = target_256.cuda(), input_256.cuda()
|
|
val_target_256, val_input_256 = val_target_256.cuda(), val_input_256.cuda()
|
|
|
|
target_128 = Variable(target_128)
|
|
input_128 = Variable(input_128)
|
|
target_256 = Variable(target_256)
|
|
input_256 = Variable(input_256)
|
|
# input = Variable(input,requires_grad=False)
|
|
# depth = Variable(depth)
|
|
ato = Variable(ato)
|
|
|
|
# Initialize VGG-16
|
|
vgg = Vgg16()
|
|
utils.init_vgg16('./models/')
|
|
vgg.load_state_dict(torch.load(os.path.join('./models/', "vgg16.weight")))
|
|
vgg.cuda()
|
|
|
|
|
|
label_d = Variable(label_d.cuda())
|
|
|
|
# get randomly sampled validation images and save it
|
|
print(len(dataloader))
|
|
val_iter = iter(valDataloader)
|
|
data_val = val_iter.next()
|
|
|
|
|
|
val_target_cpu, val_input_cpu = data_val
|
|
|
|
|
|
val_target_cpu, val_input_cpu = val_target_cpu.float().cuda(), val_input_cpu.float().cuda()
|
|
|
|
|
|
|
|
val_target.resize_as_(val_target_cpu).copy_(val_target_cpu)
|
|
val_input.resize_as_(val_input_cpu).copy_(val_input_cpu)
|
|
|
|
vutils.save_image(val_target, '%s/real_target.png' % opt.exp, normalize=True)
|
|
vutils.save_image(val_input, '%s/real_input.png' % opt.exp, normalize=True)
|
|
|
|
# pdb.set_trace()
|
|
# get optimizer
|
|
optimizerG = optim.Adam(netG.parameters(), lr = opt.lrG, betas = (opt.beta1, 0.999), weight_decay=0.00005)
|
|
# NOTE training loop
|
|
ganIterations = 0
|
|
count = 1
|
|
Best_Fs = 0
|
|
Best_epoch = 0
|
|
if opt.modeclean == 1:
|
|
Num_rn = 0
|
|
else:
|
|
Num_rn = 34
|
|
for epoch in range(1000):
|
|
if epoch%60 == 0 and epoch>0:
|
|
opt.lrG = opt.lrG/1.25
|
|
for param_group in optimizerG.param_groups:
|
|
param_group['lr'] = opt.lrG
|
|
if epoch == 200:
|
|
opt.lrG = 0.000001
|
|
for param_group in optimizerG.param_groups:
|
|
param_group['lr'] = opt.lrG
|
|
if epoch >= opt.annealStart:
|
|
adjust_learning_rate(optimizerG, opt.lrG, epoch, None, opt.annealEvery)
|
|
for set_i in range(1,5):
|
|
cl_filename ='./seg_mat_files/Images_m%d.mat'%set_i
|
|
sm_filename ='./seg_mat_files/Labels_m%d.mat'%set_i
|
|
fcl = h5py.File(cl_filename)
|
|
fsm = h5py.File(sm_filename)
|
|
clean_images = np.array(fcl['Images'])
|
|
sm_maps = np.array(fsm['Labels'])
|
|
#print(clean_images.shape)
|
|
|
|
clean_images = clean_images.transpose([0,1,3,2])
|
|
sm_maps = sm_maps.transpose([0,1,3,2])
|
|
|
|
|
|
|
|
clean_images = clean_images/255.0
|
|
clean_images = (clean_images-0.5)/0.5
|
|
sm_maps = sm_maps/(1*255.0)
|
|
sm_maps[sm_maps>0.49] = 1
|
|
sm_maps[sm_maps<=0.49] = 0
|
|
for i in range(50):
|
|
sm_map = sm_maps[i*10:i*10+10,:,:,:]
|
|
cl_image = clean_images[i*10:i*10+10,:,:,:]
|
|
#print(np.amax(sm_map))
|
|
#print(sm_map.shape)
|
|
cl_image = np.reshape(cl_image,[10,3,160,160])
|
|
sm_map = np.reshape(sm_map,[10,4,160,160])
|
|
input_cpu = Variable(torch.from_numpy(cl_image))
|
|
target_cpu = Variable(torch.from_numpy(sm_map))
|
|
x1 = int((160-opt.imageSize)/2)
|
|
y1 = int((160-opt.imageSize)/2)
|
|
input_cpu = input_cpu.numpy()
|
|
target_cpu = target_cpu.numpy()
|
|
if (random.randint(0,100)<Num_rn) :
|
|
|
|
for j in range(10):
|
|
index = random.randint(0,24000)
|
|
for k in range(4):
|
|
target_cpu[j,k,:,:]= signal.convolve(target_cpu[j,k,:,:],kernels[index,:,:],mode='same')
|
|
input_cpu[j,0,:,:]= signal.convolve(input_cpu[j,0,:,:],kernels[index,:,:],mode='same')
|
|
input_cpu[j,1,:,:]= signal.convolve(input_cpu[j,1,:,:],kernels[index,:,:],mode='same')
|
|
input_cpu[j,2,:,:]= signal.convolve(input_cpu[j,2,:,:],kernels[index,:,:],mode='same')
|
|
#input_cpu = input_cpu + (1.0/255.0)* np.random.normal(0,4,input_cpu.shape)
|
|
#input_cpu = input_cpu[:,:,x1:x1+opt.imageSize,y1:y1+opt.imageSize]
|
|
#target_cpu = target_cpu[:,:,x1:x1+opt.imageSize,y1:y1+opt.imageSize]
|
|
|
|
input_cpu = input_cpu[:,:,x1:x1+opt.imageSize,y1:y1+opt.imageSize]
|
|
target_cpu = target_cpu[:,:,x1:x1+opt.imageSize,y1:y1+opt.imageSize]
|
|
target_cpu[target_cpu>0.49]=1
|
|
target_cpu[target_cpu<=0.49]=0
|
|
input_cpu = torch.from_numpy(input_cpu)
|
|
target_cpu = torch.from_numpy(target_cpu)
|
|
target_cpu, input_cpu = target_cpu.float().cuda(), input_cpu.float().cuda()
|
|
target.data.resize_as_(target_cpu).copy_(target_cpu)
|
|
input.data.resize_as_(input_cpu).copy_(input_cpu)
|
|
input_256 = torch.nn.functional.interpolate(input,scale_factor=0.5)
|
|
target_256 = torch.nn.functional.interpolate(target,scale_factor=0.5)
|
|
|
|
x_hat, x_hat64 = netG(input,input_256)
|
|
#print(x_hat.size())
|
|
netG.zero_grad()
|
|
|
|
L_img_ = criterionCAE(x_hat, target) + 0.5*criterionCAE(x_hat64, target_256)
|
|
|
|
L_img = lambdaIMG * L_img_
|
|
|
|
if lambdaIMG != 0:
|
|
L_img.backward(retain_graph=True)
|
|
#print("came")
|
|
|
|
optimizerG.step()
|
|
ganIterations += 1
|
|
if ganIterations % (10*opt.display) == 0:
|
|
print('[%d/%d][%d/%d] L_D: %f L_img: %f L_G: %f D(x): %f D(G(z)): %f / %f'
|
|
% (epoch, opt.niter, 10*i, 2000,
|
|
L_img.data[0], L_img.data[0], L_img.data[0], L_img.data[0], L_img.data[0], L_img.data[0]))
|
|
if ganIterations % (40*opt.display) == 0:
|
|
val_batch_output = torch.zeros([16,3,128,128], dtype=torch.float32)#torch.FloatTensor([10,3,128,128]).fill_(0)
|
|
Fs = 0
|
|
for idx in range(33):
|
|
tcl_filename ='./seg_mat_files/Images_test.mat'
|
|
tsm_filename ='./seg_mat_files/Labels_test.mat'
|
|
tfcl = h5py.File(tcl_filename)
|
|
tfsm = h5py.File(tsm_filename)
|
|
tclean_images = np.array(tfcl['Images'])
|
|
tsm_maps = np.array(tfsm['Labels'])
|
|
#print(clean_images.shape)
|
|
|
|
tclean_images = tclean_images.transpose([0,1,3,2])
|
|
tsm_maps = tsm_maps.transpose([0,1,3,2])
|
|
|
|
|
|
tclean_images = tclean_images/255.0
|
|
tclean_images = (tclean_images-0.5)/0.5
|
|
tsm_maps = tsm_maps/(1*255.0)
|
|
tsm_maps[tsm_maps>0.49] = 1
|
|
tsm_maps[tsm_maps<=0.49] = 0
|
|
x1 = int((160-opt.imageSize)/2)
|
|
y1 = int((160-opt.imageSize)/2)
|
|
# print(tclean_images.shape,x1,y1,opt.imageSize)
|
|
single_img = tclean_images[idx*10:idx*10+10,:,:,:]
|
|
single_img = np.reshape(single_img,[10,3,160,160])#val_input[idx,:,:,:].unsqueeze(0)
|
|
val_inputv = Variable(torch.from_numpy(single_img), volatile=True)
|
|
yval = tsm_maps[idx*10:idx*10+10,:,:,:]
|
|
yval = np.reshape(yval,[10,4,160,160])#val_input[idx,:,:,:].unsqueeze(0)
|
|
|
|
# if epoch>=0:
|
|
# index = idx+24500
|
|
# val_inputv = val_inputv.cpu().numpy()
|
|
# val_inputv[0,0,:,:]= signal.convolve(val_inputv[0,0,:,:],kernels[index,:,:],mode='same')
|
|
# val_inputv[0,1,:,:]= signal.convolve(val_inputv[0,1,:,:],kernels[index,:,:],mode='same')
|
|
# val_inputv[0,2,:,:]= signal.convolve(val_inputv[0,2,:,:],kernels[index,:,:],mode='same')
|
|
# val_inputv = val_inputv[:,:,x1:x1+opt.imageSize,y1:y1+opt.imageSize]
|
|
# val_inputv = val_inputv + (1.0/255.0)* np.random.normal(0,4,val_inputv.shape)
|
|
# val_inputv = torch.from_numpy(val_inputv)
|
|
# val_inputv = val_inputv.float().cuda()
|
|
# else :
|
|
val_inputv = val_inputv[:,:,x1:x1+opt.imageSize,y1:y1+opt.imageSize]
|
|
yval = yval[:,:,x1:x1+opt.imageSize,y1:y1+opt.imageSize]
|
|
val_inputv = val_inputv.float().cuda()
|
|
val_inputv_256 = torch.nn.functional.interpolate(val_inputv,scale_factor=0.5)
|
|
with torch.no_grad():
|
|
x_hat_val,x_hat_val64 = netG(val_inputv,val_inputv_256)
|
|
|
|
val_batch_output[idx%16,:,:,:].copy_(x_hat_val.data[0,idx%4,:,:])
|
|
# print(val_inputv.size(),yval.min(),yval.max(),x_hat_val.min(),x_hat_val.max())
|
|
x_hat_val = x_hat_val.cpu().numpy()
|
|
x_hat_val[x_hat_val>0.5] = 1
|
|
x_hat_val[x_hat_val<=0.5] = 0
|
|
# print(val_inputv.size(),yval.min(),yval.max(),x_hat_val.min(),x_hat_val.max())
|
|
for jdx in range(10):
|
|
ttp = np.sum(yval[jdx,:,:,:]==1)
|
|
tp = np.sum((x_hat_val[jdx,:,:,:]==1) & (yval[jdx,:,:,:]==1))
|
|
fp = np.sum((x_hat_val[jdx,:,:,:] == 1) & (yval[jdx,:,:,:] == 0))
|
|
fn = np.sum((x_hat_val[jdx,:,:,:] == 0) & (yval[jdx,:,:,:] == 1))
|
|
tot = tp+fp
|
|
tot2 = tp+fn
|
|
if tot==0 or tot2 ==0 or tp==0:
|
|
|
|
f1_s = 1
|
|
else:
|
|
precision = tp / (tp + fp);
|
|
recall = tp / (tp + fn);
|
|
f1_s = (2 * precision * recall) / (precision + recall);
|
|
#print(f1_s)
|
|
|
|
Fs = Fs + f1_s
|
|
# print(f1_s)
|
|
|
|
|
|
|
|
Fs = Fs/330
|
|
if Best_Fs<Fs:
|
|
Best_Fs = Fs
|
|
Best_epoch = epoch
|
|
print("Best epoch: %d"%(Best_epoch))
|
|
print("Best Fs: %f\t Fs:%f"%(Best_Fs,Fs))
|
|
if ganIterations % (40*opt.display) == 0:
|
|
vutils.save_image(val_batch_output, '%s/generated_epoch_iter%08d.png' % \
|
|
(opt.exp, ganIterations), normalize=True, scale_each=False)
|
|
|
|
sys.stdout.flush()
|
|
trainLogger.write('%d\t%f\t%f\t%f\t%f\t%f\t%f\n' % \
|
|
(i, L_img.data[0], L_img.data[0], L_img.data[0], L_img.data[0], L_img.data[0], L_img.data[0]))
|
|
trainLogger.flush()
|
|
torch.save(netG.state_dict(), '%s/SMaps_%d.pth' % (opt.exp, epoch))
|
|
|
|
trainLogger.close()
|
|
|
|
|