Python _netW示例

示例#1

文件： train_G.py 项目： wenchaodudu/visDial.pytorch

####################################################################################
# Build the Model
####################################################################################

vocab_size = dataset.vocab_size
ques_length = dataset.ques_length
ans_length = dataset.ans_length + 1
his_length = dataset.ques_length + dataset.ans_length
itow = dataset.itow
img_feat_size = opt.conv_feat_size

netE = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout,
             img_feat_size)

netW = model._netW(vocab_size, opt.ninp, opt.dropout)
netG = _netG(opt.model, vocab_size, opt.ninp, opt.nhid, opt.nlayers,
             opt.dropout)
critG = model.LMCriterion()
sampler = model.gumbel_sampler()

if opt.cuda:
    netW.cuda()
    netE.cuda()
    netG.cuda()
    critG.cuda()
    sampler.cuda()

if opt.model_path != '':
    netW.load_state_dict(checkpoint['netW'])
    netE.load_state_dict(checkpoint['netE'])

示例#2

文件： train_all.py 项目： smit14/visDial.pytorch

                                             num_workers=int(opt.workers))

####################################################################################
# Build the Model
####################################################################################
vocab_size = dataset.vocab_size
ques_length = dataset.ques_length
ans_length = dataset.ans_length + 1
his_length = dataset.ans_length + dataset.ques_length
itow = dataset.itow
img_feat_size = 512

print('init Discriminator model...')
netE_d = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout,
               img_feat_size)
netW_d = model._netW(vocab_size, opt.ninp, opt.dropout)
netD = model._netD(opt.model, opt.ninp, opt.nhid, opt.nlayers, vocab_size,
                   opt.dropout)
critD = model.nPairLoss(opt.ninp, opt.margin)

if opt.model_path_D != '':
    print('Loading Discriminator model...')
    netW_d.load_state_dict(checkpoint_D['netW'])
    netE_d.load_state_dict(checkpoint_D['netE'])
    netD.load_state_dict(checkpoint_D['netD'])

print('init Generative model...')
netE_g = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout,
               img_feat_size)
netW_g = model._netW(vocab_size, opt.ninp, opt.dropout)
netG = _netG(opt.model, vocab_size, opt.ninp, opt.nhid, opt.nlayers,

示例#3

文件： train_D.py 项目： AashishV/visDial.pytorch

dataloader_val = torch.utils.data.DataLoader(dataset_val, batch_size=1,
                                         shuffle=False, num_workers=int(opt.workers))

####################################################################################
# Build the Model
####################################################################################
n_neg = opt.negative_sample
vocab_size = dataset.vocab_size
ques_length = dataset.ques_length
ans_length = dataset.ans_length + 1
his_length = dataset.ans_length + dataset.ques_length
itow = dataset.itow
img_feat_size = 512

netE = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout, img_feat_size)
netW = model._netW(vocab_size, opt.ninp, opt.dropout)
netD = model._netD(opt.model, opt.ninp, opt.nhid, opt.nlayers, vocab_size, opt.dropout)
critD =model.nPairLoss(opt.ninp, opt.margin)

if opt.model_path != '': # load the pre-trained model.
    netW.load_state_dict(checkpoint['netW'])
    netE.load_state_dict(checkpoint['netE'])
    netD.load_state_dict(checkpoint['netD'])

if opt.cuda: # ship to cuda, if has GPU
    netW.cuda(), netE.cuda(),
    netD.cuda(), critD.cuda()

####################################################################################
# training model
####################################################################################

示例#4

                                             num_workers=int(opt.workers))
####################################################################################
# Build the Model
####################################################################################

n_words = dataset_val.vocab_size
ques_length = dataset_val.ques_length
ans_length = dataset_val.ans_length + 1
his_length = ques_length + dataset_val.ans_length
itow = dataset_val.itow
img_feat_size = 512

netE = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout,
             img_feat_size)

netW = model._netW(n_words, opt.ninp, opt.dropout)
netD = model._netD(opt.model, opt.ninp, opt.nhid, opt.nlayers, n_words,
                   opt.dropout)
critD = model.nPairLoss(opt.nhid, 2)

netW.load_state_dict(checkpoint['netW'])
netE.load_state_dict(checkpoint['netE'])
netD.load_state_dict(checkpoint['netD'])
print('Loading model Success!')

if opt.cuda:  # ship to cuda, if has GPU
    netW.cuda(), netE.cuda(), netD.cuda()
    critD.cuda()

n_neg = 100
####################################################################################

示例#5

文件： eval_G_DIS.py 项目： AashishV/visDial.pytorch

dataloader_val = torch.utils.data.DataLoader(dataset_val, batch_size=opt.batchSize,
                                         shuffle=False, num_workers=int(opt.workers))
####################################################################################
# Build the Model
####################################################################################

n_words = dataset_val.vocab_size
ques_length = dataset_val.ques_length
ans_length = dataset_val.ans_length + 1
his_length = ques_length+dataset_val.ans_length
itow = dataset_val.itow
img_feat_size = 512

netE = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout, img_feat_size)
netW = model._netW(n_words, opt.ninp, opt.dropout)
netG = _netG(opt.model, n_words, opt.ninp, opt.nhid, opt.nlayers, opt.dropout)
critG = model.LMCriterion()
sampler = model.gumbel_sampler()


if opt.cuda:
    netW.cuda()
    netE.cuda()
    netG.cuda()
    critG.cuda()
    sampler.cuda()

if opt.model_path != '':
    netW.load_state_dict(checkpoint['netW_g'])
    netE.load_state_dict(checkpoint['netE_g'])

示例#6

                                             shuffle=False,
                                             num_workers=int(opt.workers))

####################################################################################
# Build the Model
####################################################################################
n_neg = opt.negative_sample
vocab_size = dataset.vocab_size
ques_length = dataset.ques_length
#ans_length = dataset.ans_length + 1
ans_length = dataset.ans_length
his_length = dataset.ans_length + dataset.ques_length
itow = dataset.itow

netE = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout)
netW = model._netW(vocab_size, opt.ninp, opt.dropout, dataset.pretrained_wemb)
netD = model._netD(opt.model, opt.ninp, opt.nhid, opt.nlayers, vocab_size,
                   opt.dropout)
critD = model.nPairLoss(opt.ninp, opt.margin, 0.1)

if opt.model_path != '':  # load the pre-trained model.
    netW.load_state_dict(checkpoint['netW'])
    netE.load_state_dict(checkpoint['netE'])
    netD.load_state_dict(checkpoint['netD'])

if opt.cuda:  # ship to cuda, if has GPU
    netW.cuda(), netE.cuda(),
    netD.cuda(), critD.cuda()


####################################################################################

示例#7

文件： train_all.py 项目： AashishV/visDial.pytorch

dataloader_val = torch.utils.data.DataLoader(dataset_val, batch_size=10,
                                         shuffle=False, num_workers=int(opt.workers))

####################################################################################
# Build the Model
####################################################################################
vocab_size = dataset.vocab_size
ques_length = dataset.ques_length
ans_length = dataset.ans_length + 1
his_length = dataset.ans_length + dataset.ques_length
itow = dataset.itow
img_feat_size = 512

print('init Discriminator model...')
netE_d = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout, img_feat_size)
netW_d = model._netW(vocab_size, opt.ninp, opt.dropout)
netD = model._netD(opt.model, opt.ninp, opt.nhid, opt.nlayers, vocab_size, opt.dropout)
critD =model.nPairLoss(opt.ninp, opt.margin)

if opt.model_path_D != '' :
    print('Loading Discriminator model...')
    netW_d.load_state_dict(checkpoint_D['netW'])
    netE_d.load_state_dict(checkpoint_D['netE'])
    netD.load_state_dict(checkpoint_D['netD'])

print('init Generative model...')
netE_g = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout, img_feat_size)
netW_g = model._netW(vocab_size, opt.ninp, opt.dropout)
netG = _netG(opt.model, vocab_size, opt.ninp, opt.nhid, opt.nlayers, opt.dropout)
sampler = model.gumbel_sampler()
critG = model.G_loss(opt.ninp)

示例#8

文件： evaluate.py 项目： wh0330/CAG_VisDial

                                             shuffle=False,
                                             num_workers=int(opt.workers))
####################################################################################
# Build the Model
####################################################################################

n_words = dataset_val.vocab_size
ques_length = dataset_val.ques_length
ans_length = dataset_val.ans_length + 1
his_length = ques_length + dataset_val.ans_length
itow = dataset_val.itow
img_feat_size = 512

netE = _netE(opt.model, opt.ninp, opt.nhid, opt.nlayers, opt.dropout,
             img_feat_size)
netW = model._netW(n_words, opt.ninp, opt.dropout, dataset_val.pretrained_wemb)
netD = model._netD(opt.model, opt.ninp, opt.nhid, opt.nlayers, n_words,
                   opt.dropout)
#critD = model.nPairLoss(opt.nhid, 2)

if opt.model_path != '':  # load the pre-trained model.
    netW.load_state_dict(checkpoint['netW'])
    netE.load_state_dict(checkpoint['netE'])
    netD.load_state_dict(checkpoint['netD'])
    print('Loading model Success!')

if opt.cuda:  # ship to cuda, if has GPU
    netW.cuda(), netE.cuda(), netD.cuda()
#    critD.cuda()

n_neg = 100