def __init__(self, opt):
self.log = create_logger(__name__, silent=False, to_disk=True,
log_file=cfg.log_filename if cfg.if_test
else [cfg.log_filename, cfg.save_root + 'log.txt'])
self.sig = Signal(cfg.signal_file)
self.opt = opt
self.show_config()
self.clas = None
# load dictionary
self.word2idx_dict, self.idx2word_dict = load_dict(cfg.dataset)
# Dataloader
try:
self.train_data = GenDataIter(cfg.train_data)
self.test_data = GenDataIter(cfg.test_data, if_test_data=True)
except:
pass
try:
self.train_data_list = [GenDataIter(cfg.cat_train_data.format(i)) for i in range(cfg.k_label)]
self.test_data_list = [GenDataIter(cfg.cat_test_data.format(i), if_test_data=True) for i in
range(cfg.k_label)]
self.clas_data_list = [GenDataIter(cfg.cat_test_data.format(str(i)), if_test_data=True) for i in
range(cfg.k_label)]
self.train_samples_list = [self.train_data_list[i].target for i in range(cfg.k_label)]
self.clas_samples_list = [self.clas_data_list[i].target for i in range(cfg.k_label)]
except:
pass
# Criterion
self.mle_criterion = nn.NLLLoss()
self.dis_criterion = nn.CrossEntropyLoss()
self.clas_criterion = nn.CrossEntropyLoss()
# Optimizer
self.clas_opt = None
# Metrics
self.bleu = BLEU('BLEU', gram=[2, 3, 4, 5], if_use=cfg.use_bleu)
self.nll_gen = NLL('NLL_gen', if_use=cfg.use_nll_gen, gpu=cfg.CUDA)
self.nll_div = NLL('NLL_div', if_use=cfg.use_nll_div, gpu=cfg.CUDA)
self.self_bleu = BLEU('Self-BLEU', gram=[2, 3, 4], if_use=cfg.use_self_bleu)
self.clas_acc = ACC(if_use=cfg.use_clas_acc)
self.ppl = PPL(self.train_data, self.test_data, n_gram=5, if_use=cfg.use_ppl)
self.all_metrics = [self.bleu, self.nll_gen, self.nll_div, self.self_bleu, self.ppl]
def create_many_oracle(from_a, to_b, num=1, save_path='../pretrain/'):
for i in range(num):
while True:
oracle = Oracle(cfg.gen_embed_dim,
cfg.gen_hidden_dim,
cfg.vocab_size,
cfg.max_seq_len,
cfg.padding_idx,
gpu=cfg.CUDA)
if cfg.CUDA:
oracle = oracle.cuda()
big_samples = oracle.sample(cfg.samples_num, 8 * cfg.batch_size)
small_samples = oracle.sample(cfg.samples_num // 2,
8 * cfg.batch_size)
oracle_data = GenDataIter(big_samples)
mle_criterion = nn.NLLLoss()
groud_truth = NLL.cal_nll(oracle, oracle_data.loader,
mle_criterion)
if from_a <= groud_truth <= to_b:
print('save ground truth: ', groud_truth)
prefix = 'oracle_lstm'
torch.save(oracle.state_dict(),
save_path + '{}.pt'.format(prefix))
torch.save(
big_samples, save_path +
'{}_samples_{}.pt'.format(prefix, cfg.samples_num))
torch.save(
small_samples, save_path +
'{}_samples_{}.pt'.format(prefix, cfg.samples_num // 2))
break
def create_multi_oracle(number):
for i in range(number):
print('Creating Oracle %d...' % i)
oracle = Oracle(cfg.gen_embed_dim,
cfg.gen_hidden_dim,
cfg.vocab_size,
cfg.max_seq_len,
cfg.padding_idx,
gpu=cfg.CUDA)
if cfg.CUDA:
oracle = oracle.cuda()
large_samples = oracle.sample(cfg.samples_num, 4 * cfg.batch_size)
small_samples = oracle.sample(cfg.samples_num // 2, 4 * cfg.batch_size)
torch.save(oracle.state_dict(),
cfg.multi_oracle_state_dict_path.format(i))
torch.save(large_samples,
cfg.multi_oracle_samples_path.format(i, cfg.samples_num))
torch.save(
small_samples,
cfg.multi_oracle_samples_path.format(i, cfg.samples_num // 2))
oracle_data = GenDataIter(large_samples)
mle_criterion = nn.NLLLoss()
groud_truth = NLL.cal_nll(oracle, oracle_data.loader, mle_criterion)
print('Oracle %d Groud Truth: %.4f' % (i, groud_truth))
def create_oracle():
"""Create a new Oracle model and Oracle's samples"""
import config as cfg
from models.Oracle import Oracle
print('Creating Oracle...')
oracle = Oracle(cfg.gen_embed_dim,
cfg.gen_hidden_dim,
cfg.vocab_size,
cfg.max_seq_len,
cfg.padding_idx,
gpu=cfg.CUDA)
if cfg.CUDA:
oracle = oracle.cuda()
torch.save(oracle.state_dict(), cfg.oracle_state_dict_path)
big_samples = oracle.sample(cfg.samples_num, 4 * cfg.batch_size)
# large
torch.save(big_samples, cfg.oracle_samples_path.format(cfg.samples_num))
# small
torch.save(oracle.sample(cfg.samples_num // 2, 4 * cfg.batch_size),
cfg.oracle_samples_path.format(cfg.samples_num // 2))
oracle_data = GenDataIter(big_samples)
mle_criterion = nn.NLLLoss()
groud_truth = NLL.cal_nll(oracle, oracle_data.loader, mle_criterion)
print('NLL_Oracle Groud Truth: %.4f' % groud_truth)
def __init__(self, opt):
self.log = create_logger(__name__, silent=False, to_disk=True,
log_file=cfg.log_filename if cfg.if_test
else [cfg.log_filename, cfg.save_root + 'log.txt'])
self.sig = Signal(cfg.signal_file)
self.opt = opt
# oracle, generator, discriminator
self.oracle = Oracle(cfg.gen_embed_dim, cfg.gen_hidden_dim, cfg.vocab_size, cfg.max_seq_len,
cfg.padding_idx, gpu=cfg.CUDA)
self.oracle_list = [Oracle(cfg.gen_embed_dim, cfg.gen_hidden_dim, cfg.vocab_size, cfg.max_seq_len,
cfg.padding_idx, gpu=cfg.CUDA) for _ in range(cfg.k_label)]
self.dis = None
self.clas = None
self.show_config()
self.check_oracle() # Create Oracle models if not exist
# DataLoader
self.oracle_samples = torch.load(cfg.oracle_samples_path.format(cfg.samples_num))
self.oracle_samples_list = [torch.load(cfg.multi_oracle_samples_path.format(i, cfg.samples_num))
for i in range(cfg.k_label)]
self.oracle_data = GenDataIter(self.oracle_samples)
self.oracle_data_list = [GenDataIter(self.oracle_samples_list[i]) for i in range(cfg.k_label)]
# Criterion
self.mle_criterion = nn.NLLLoss()
self.dis_criterion = nn.CrossEntropyLoss()
# Metrics
self.nll_oracle = NLL('NLL_oracle', if_use=cfg.use_nll_oracle, gpu=cfg.CUDA)
self.nll_gen = NLL('NLL_gen', if_use=cfg.use_nll_gen, gpu=cfg.CUDA)
self.nll_div = NLL('NLL_div', if_use=cfg.use_nll_div, gpu=cfg.CUDA)
self.all_metrics = [self.nll_oracle, self.nll_gen, self.nll_div]
def evaluation(self, eval_type):
"""Evaluation all children, update child score. Note that the eval data should be the same"""
eval_samples = self.gen.sample(cfg.eval_b_num * cfg.batch_size,
cfg.max_bn * cfg.batch_size)
gen_data = GenDataIter(eval_samples)
# Fd
if cfg.lambda_fd != 0:
Fd = NLL.cal_nll(self.gen, gen_data.loader,
self.mle_criterion) # NLL_div
else:
Fd = 0
if eval_type == 'standard':
Fq = self.eval_d_out_fake.mean().cpu().item()
elif eval_type == 'rsgan':
g_loss, d_loss = get_losses(self.eval_d_out_real,
self.eval_d_out_fake, 'rsgan')
Fq = d_loss.item()
elif eval_type == 'nll':
if cfg.lambda_fq != 0:
Fq = -NLL.cal_nll(self.oracle, gen_data.loader,
self.mle_criterion) # NLL_Oracle
else:
Fq = 0
elif eval_type == 'Ra':
g_loss = torch.sigmoid(self.eval_d_out_fake -
torch.mean(self.eval_d_out_real)).sum()
Fq = g_loss.item()
else:
raise NotImplementedError("Evaluation '%s' is not implemented" %
eval_type)
score = cfg.lambda_fq * Fq + cfg.lambda_fd * Fd
return Fq, Fd, score
def evaluation(self, eval_type):
"""Evaluation all children, update child score. Note that the eval data should be the same"""
eval_samples = [
self.gen.sample(cfg.eval_b_num * cfg.batch_size,
cfg.max_bn * cfg.batch_size,
label_i=i) for i in range(cfg.k_label)
]
# Fd
if cfg.lambda_fd != 0:
nll_div = []
for label_i in range(cfg.k_label):
gen_data = GenDataIter(eval_samples[label_i])
nll_div.append(
NLL.cal_nll_with_label(self.gen, gen_data.loader, label_i,
self.mle_criterion))
Fd = sum(nll_div)
else:
Fd = 0
# Fq
if 'bleu' in eval_type:
bleu_score = []
for i in range(cfg.k_label):
bleu_score.append(
self.bleu[i].get_score(given_gram=int(eval_type[-1])))
Fq = sum(bleu_score)
elif 'Ra' in eval_type:
g_loss = 0
for i in range(cfg.k_label):
g_loss += torch.sigmoid(
self.eval_d_out_fake[i] -
torch.mean(self.eval_d_out_real[i])).sum()
Fq = g_loss.item()
else:
raise NotImplementedError("Evaluation '%s' is not implemented" %
eval_type)
score = cfg.lambda_fq * Fq + cfg.lambda_fd * Fd
return Fq, Fd, score
def evaluation(self, eval_type):
"""Evaluation all children, update child score. Note that the eval data should be the same"""
eval_samples = self.gen.sample(cfg.eval_b_num * cfg.batch_size,
cfg.max_bn * cfg.batch_size)
gen_data = GenDataIter(eval_samples)
# Fd
if cfg.lambda_fd != 0:
Fd = NLL.cal_nll(self.gen, gen_data.loader,
self.mle_criterion) # NLL_div
else:
Fd = 0
# Fq
if eval_type == 'standard':
Fq = self.eval_d_out_fake.mean().cpu().item()
elif eval_type == 'rsgan':
g_loss, d_loss = get_losses(self.eval_d_out_real,
self.eval_d_out_fake, 'rsgan')
Fq = d_loss.item()
elif 'bleu' in eval_type:
self.bleu.reset(
test_text=tensor_to_tokens(eval_samples, self.idx2word_dict))
if cfg.lambda_fq != 0:
Fq = self.bleu.get_score(given_gram=int(eval_type[-1]))
else:
Fq = 0
elif 'Ra' in eval_type:
g_loss = torch.sigmoid(self.eval_d_out_fake -
torch.mean(self.eval_d_out_real)).sum()
Fq = g_loss.item()
else:
raise NotImplementedError("Evaluation '%s' is not implemented" %
eval_type)
score = cfg.lambda_fq * Fq + cfg.lambda_fd * Fd
return Fq, Fd, score
def create_specific_oracle(from_a, to_b, num=1, save_path='../pretrain/'):
for i in range(num):
while True:
oracle = Oracle(cfg.gen_embed_dim,
cfg.gen_hidden_dim,
cfg.vocab_size,
cfg.max_seq_len,
cfg.padding_idx,
gpu=cfg.CUDA)
if cfg.CUDA:
oracle = oracle.cuda()
big_samples = oracle.sample(cfg.samples_num, 8 * cfg.batch_size)
small_samples = oracle.sample(cfg.samples_num // 2,
8 * cfg.batch_size)
oracle_data = GenDataIter(big_samples)
mle_criterion = nn.NLLLoss()
groud_truth = NLL.cal_nll(oracle, oracle_data.loader,
mle_criterion)
if from_a <= groud_truth <= to_b:
dir_path = save_path + 'oracle_data_gt{:.2f}_{}'.format(
groud_truth, strftime("%m%d_%H%M%S", localtime()))
if not os.path.exists(dir_path):
os.mkdir(dir_path)
print('save ground truth: ', groud_truth)
# prefix = 'oracle{}_lstm_gt{:.2f}_{}'.format(i, groud_truth, strftime("%m%d", localtime()))
prefix = dir_path + '/oracle_lstm'
torch.save(oracle.state_dict(), '{}.pt'.format(prefix))
torch.save(big_samples,
'{}_samples_{}.pt'.format(prefix, cfg.samples_num))
torch.save(
small_samples,
'{}_samples_{}.pt'.format(prefix, cfg.samples_num // 2))
break
class BasicInstructor:
def __init__(self, opt):
self.log = create_logger(__name__, silent=False, to_disk=True,
log_file=cfg.log_filename if cfg.if_test
else [cfg.log_filename, cfg.save_root + 'log.txt'])
self.sig = Signal(cfg.signal_file)
self.opt = opt
# oracle, generator, discriminator
self.oracle = Oracle(cfg.gen_embed_dim, cfg.gen_hidden_dim, cfg.vocab_size, cfg.max_seq_len,
cfg.padding_idx, gpu=cfg.CUDA)
self.oracle_list = [Oracle(cfg.gen_embed_dim, cfg.gen_hidden_dim, cfg.vocab_size, cfg.max_seq_len,
cfg.padding_idx, gpu=cfg.CUDA) for _ in range(cfg.k_label)]
self.dis = None
self.clas = None
self.show_config()
self.check_oracle() # Create Oracle models if not exist
# DataLoader
self.oracle_samples = torch.load(cfg.oracle_samples_path.format(cfg.samples_num))
self.oracle_samples_list = [torch.load(cfg.multi_oracle_samples_path.format(i, cfg.samples_num))
for i in range(cfg.k_label)]
self.oracle_data = GenDataIter(self.oracle_samples)
self.oracle_data_list = [GenDataIter(self.oracle_samples_list[i]) for i in range(cfg.k_label)]
# Criterion
self.mle_criterion = nn.NLLLoss()
self.dis_criterion = nn.CrossEntropyLoss()
# Metrics
self.nll_oracle = NLL('NLL_oracle', if_use=cfg.use_nll_oracle, gpu=cfg.CUDA)
self.nll_gen = NLL('NLL_gen', if_use=cfg.use_nll_gen, gpu=cfg.CUDA)
self.nll_div = NLL('NLL_div', if_use=cfg.use_nll_div, gpu=cfg.CUDA)
self.all_metrics = [self.nll_oracle, self.nll_gen, self.nll_div]
def _run(self):
print('Nothing to run in Basic Instructor!')
pass
def _test(self):
pass
def init_model(self):
if cfg.oracle_pretrain:
if not os.path.exists(cfg.oracle_state_dict_path):
create_oracle()
self.oracle.load_state_dict(torch.load(cfg.oracle_state_dict_path))
if cfg.dis_pretrain:
self.log.info(
'Load pretrained discriminator: {}'.format(cfg.pretrained_dis_path))
self.dis.load_state_dict(torch.load(cfg.pretrained_dis_path))
if cfg.gen_pretrain:
self.log.info('Load MLE pretrained generator gen: {}'.format(cfg.pretrained_gen_path))
self.gen.load_state_dict(torch.load(cfg.pretrained_gen_path, map_location='cuda:{}'.format(cfg.device)))
if cfg.CUDA:
self.oracle = self.oracle.cuda()
self.gen = self.gen.cuda()
self.dis = self.dis.cuda()
def train_gen_epoch(self, model, data_loader, criterion, optimizer):
total_loss = 0
for i, data in enumerate(data_loader):
inp, target = data['input'], data['target']
'''
print("inp.shape = ",inp.shape) -> [64,20]
if (inp.numpy()[0][1:] == target.numpy()[0][:-1]).all:
print("yes") 没错
exit()
'''
if cfg.CUDA:
inp, target = inp.cuda(), target.cuda()
hidden = model.init_hidden(data_loader.batch_size)
pred = model.forward(inp, hidden) #seqGAN:(batch_size * seq_len) * vocab_size
loss = criterion(pred, target.view(-1)) #seqGAN:self.mle_criterion = nn.NLLLoss()
self.optimize(optimizer, loss, model)
total_loss += loss.item()
return total_loss / len(data_loader)
def train_dis_epoch(self, model, data_loader, criterion, optimizer):
total_loss = 0
total_acc = 0
total_num = 0
for i, data in enumerate(data_loader):
inp, target = data['input'], data['target']
if cfg.CUDA:
inp, target = inp.cuda(), target.cuda()
pred = model.forward(inp)
loss = criterion(pred, target)
self.optimize(optimizer, loss, model)
total_loss += loss.item()
total_acc += torch.sum((pred.argmax(dim=-1) == target)).item()
total_num += inp.size(0)
total_loss /= len(data_loader)
total_acc /= total_num
return total_loss, total_acc
@staticmethod
def eval_dis(model, data_loader, criterion):
total_loss = 0
total_acc = 0
total_num = 0
with torch.no_grad():
for i, data in enumerate(data_loader):
inp, target = data['input'], data['target']
if cfg.CUDA:
inp, target = inp.cuda(), target.cuda()
pred = model.forward(inp)
loss = criterion(pred, target)
total_loss += loss.item()
total_acc += torch.sum((pred.argmax(dim=-1) == target)).item()
total_num += inp.size(0)
total_loss /= len(data_loader)
total_acc /= total_num
return total_loss, total_acc
@staticmethod
def optimize_multi(opts, losses):
for i, (opt, loss) in enumerate(zip(opts, losses)):
opt.zero_grad()
loss.backward(retain_graph=True if i < len(opts) - 1 else False)
opt.step()
@staticmethod
def optimize(opt, loss, model=None, retain_graph=False):
opt.zero_grad()
loss.backward(retain_graph=retain_graph)
if model is not None:
torch.nn.utils.clip_grad_norm_(model.parameters(), cfg.clip_norm)
opt.step()
def show_config(self):
"""Show parser parameters settings"""
self.log.info(100 * '=')
self.log.info('> training arguments:')
for arg in vars(self.opt):
self.log.info('>>> {0}: {1}'.format(arg, getattr(self.opt, arg)))
self.log.info(100 * '=')
def cal_metrics(self, fmt_str=False):
"""
Calculate metrics
:param fmt_str: if return format string for logging
"""
with torch.no_grad():
# Prepare data for evaluation
gen_data = GenDataIter(self.gen.sample(cfg.samples_num, 4 * cfg.batch_size))
# Reset metrics
self.nll_oracle.reset(self.oracle, gen_data.loader)
self.nll_gen.reset(self.gen, self.oracle_data.loader)
self.nll_div.reset(self.gen, gen_data.loader)
if fmt_str:
return ', '.join(['%s = %s' % (metric.get_name(), metric.get_score()) for metric in self.all_metrics])
else:
return [metric.get_score() for metric in self.all_metrics]
def cal_metrics_with_label(self, label_i):
assert type(label_i) == int, 'missing label'
with torch.no_grad():
# Prepare data for evaluation
eval_samples = self.gen.sample(cfg.samples_num, 8 * cfg.batch_size, label_i=label_i)
gen_data = GenDataIter(eval_samples)
# Reset metrics
self.nll_oracle.reset(self.oracle_list[label_i], gen_data.loader, label_i)
self.nll_gen.reset(self.gen, self.oracle_data_list[label_i].loader, label_i)
self.nll_div.reset(self.gen, gen_data.loader, label_i)
return [metric.get_score() for metric in self.all_metrics]
def comb_metrics(self, fmt_str=False):
all_scores = [self.cal_metrics_with_label(label_i) for label_i in range(cfg.k_label)]
all_scores = np.array(all_scores).T.tolist() # each row for each metric
if fmt_str:
return ', '.join(['%s = %s' % (metric.get_name(), score)
for (metric, score) in zip(self.all_metrics, all_scores)])
return all_scores
def _save(self, phase, epoch):
"""Save model state dict and generator's samples"""
if phase != 'ADV':
torch.save(self.gen.state_dict(), cfg.save_model_root + 'gen_{}_{:05d}.pt'.format(phase, epoch))
save_sample_path = cfg.save_samples_root + 'samples_{}_{:05d}.txt'.format(phase, epoch)
samples = self.gen.sample(cfg.batch_size, cfg.batch_size)
write_tensor(save_sample_path, samples)
def update_temperature(self, i, N):
self.gen.temperature.data = torch.Tensor([get_fixed_temperature(cfg.temperature, i, N, cfg.temp_adpt)])
if cfg.CUDA:
self.gen.temperature.data = self.gen.temperature.data.cuda()
def check_oracle(self):
if not cfg.oracle_pretrain:
create_oracle()
create_multi_oracle(cfg.k_label)
# General text generation Oracle model
if not os.path.exists(cfg.oracle_samples_path.format(cfg.samples_num)) or not cfg.oracle_pretrain:
create_oracle()
# Category text generation Oracle models
for i in range(cfg.k_label):
if not os.path.exists(cfg.multi_oracle_samples_path.format(i, cfg.samples_num)):
create_multi_oracle(cfg.k_label)
break
# Load Oracle state dict
self.oracle.load_state_dict(torch.load(cfg.oracle_state_dict_path))
for i in range(cfg.k_label):
oracle_path = cfg.multi_oracle_state_dict_path.format(i)
self.oracle_list[i].load_state_dict(torch.load(oracle_path))
class BasicInstructor:
def __init__(self, opt):
self.log = create_logger(__name__,
silent=False,
to_disk=True,
log_file=cfg.log_filename if cfg.if_test else
[cfg.log_filename, cfg.save_root + 'log.txt'])
self.sig = Signal(cfg.signal_file)
self.opt = opt
self.show_config()
self.clas = None
# load dictionary
self.word2idx_dict, self.idx2word_dict = load_dict(cfg.dataset)
# Dataloader
try:
self.train_data = GenDataIter(cfg.train_data)
self.test_data = GenDataIter(cfg.test_data, if_test_data=True)
except:
pass
try:
self.train_data_list = [
GenDataIter(cfg.cat_train_data.format(i))
for i in range(cfg.k_label)
]
self.test_data_list = [
GenDataIter(cfg.cat_test_data.format(i), if_test_data=True)
for i in range(cfg.k_label)
]
self.clas_data_list = [
GenDataIter(cfg.cat_test_data.format(str(i)),
if_test_data=True) for i in range(cfg.k_label)
]
self.train_samples_list = [
self.train_data_list[i].target for i in range(cfg.k_label)
]
self.clas_samples_list = [
self.clas_data_list[i].target for i in range(cfg.k_label)
]
except:
pass
# Criterion
self.mle_criterion = nn.NLLLoss()
self.dis_criterion = nn.CrossEntropyLoss()
self.clas_criterion = nn.CrossEntropyLoss()
# Optimizer
self.clas_opt = None
# Metrics
self.bleu = BLEU('BLEU', gram=[2, 3, 4, 5], if_use=cfg.use_bleu)
self.nll_gen = NLL('NLL_gen', if_use=cfg.use_nll_gen, gpu=cfg.CUDA)
self.nll_div = NLL('NLL_div', if_use=cfg.use_nll_div, gpu=cfg.CUDA)
self.self_bleu = BLEU('Self-BLEU',
gram=[2, 3, 4],
if_use=cfg.use_self_bleu)
self.clas_acc = ACC(if_use=cfg.use_clas_acc)
self.ppl = PPL(self.train_data,
self.test_data,
n_gram=5,
if_use=cfg.use_ppl)
self.all_metrics = [
self.bleu, self.nll_gen, self.nll_div, self.self_bleu, self.ppl
]
def _run(self):
print('Nothing to run in Basic Instructor!')
pass
def _test(self):
pass
def init_model(self):
if cfg.dis_pretrain:
self.log.info('Load pre-trained discriminator: {}'.format(
cfg.pretrained_dis_path))
self.dis.load_state_dict(torch.load(cfg.pretrained_dis_path))
if cfg.gen_pretrain:
self.log.info('Load MLE pre-trained generator: {}'.format(
cfg.pretrained_gen_path))
self.gen.load_state_dict(torch.load(cfg.pretrained_gen_path))
if cfg.CUDA:
self.gen = self.gen.cuda()
self.dis = self.dis.cuda()
def train_gen_epoch(self, model, data_loader, criterion, optimizer):
total_loss = 0
for i, data in enumerate(data_loader):
inp, target = data['input'], data['target']
if cfg.CUDA:
inp, target = inp.cuda(), target.cuda()
hidden = model.init_hidden(data_loader.batch_size)
pred = model.forward(inp, hidden)
loss = criterion(pred, target.view(-1))
self.optimize(optimizer, loss, model)
total_loss += loss.item()
return total_loss / len(data_loader)
def train_dis_epoch(self, model, data_loader, criterion, optimizer):
total_loss = 0
total_acc = 0
total_num = 0
for i, data in enumerate(data_loader):
inp, target = data['input'], data['target']
if cfg.CUDA:
inp, target = inp.cuda(), target.cuda()
pred = model.forward(inp)
loss = criterion(pred, target)
self.optimize(optimizer, loss, model)
total_loss += loss.item()
total_acc += torch.sum((pred.argmax(dim=-1) == target)).item()
total_num += inp.size(0)
total_loss /= len(data_loader)
total_acc /= total_num
return total_loss, total_acc
def train_classifier(self, epochs):
"""
Classifier for calculating the classification accuracy metric of category text generation.
Note: the train and test data for the classifier is opposite to the generator.
Because the classifier is to calculate the classification accuracy of the generated samples
where are trained on self.train_samples_list.
Since there's no test data in synthetic data (oracle data), the synthetic data experiments
doesn't need a classifier.
"""
import copy
# Prepare data for Classifier
clas_data = CatClasDataIter(self.clas_samples_list)
eval_clas_data = CatClasDataIter(self.train_samples_list)
max_acc = 0
best_clas = None
for epoch in range(epochs):
c_loss, c_acc = self.train_dis_epoch(self.clas, clas_data.loader,
self.clas_criterion,
self.clas_opt)
_, eval_acc = self.eval_dis(self.clas, eval_clas_data.loader,
self.clas_criterion)
if eval_acc > max_acc:
best_clas = copy.deepcopy(
self.clas.state_dict()) # save the best classifier
max_acc = eval_acc
self.log.info(
'[PRE-CLAS] epoch %d: c_loss = %.4f, c_acc = %.4f, eval_acc = %.4f, max_eval_acc = %.4f',
epoch, c_loss, c_acc, eval_acc, max_acc)
self.clas.load_state_dict(
copy.deepcopy(best_clas)) # Reload the best classifier
@staticmethod
def eval_dis(model, data_loader, criterion):
total_loss = 0
total_acc = 0
total_num = 0
with torch.no_grad():
for i, data in enumerate(data_loader):
inp, target = data['input'], data['target']
if cfg.CUDA:
inp, target = inp.cuda(), target.cuda()
pred = model.forward(inp)
loss = criterion(pred, target)
total_loss += loss.item()
total_acc += torch.sum((pred.argmax(dim=-1) == target)).item()
total_num += inp.size(0)
total_loss /= len(data_loader)
total_acc /= total_num
return total_loss, total_acc
@staticmethod
def optimize_multi(opts, losses):
for i, (opt, loss) in enumerate(zip(opts, losses)):
opt.zero_grad()
loss.backward(retain_graph=True if i < len(opts) - 1 else False)
opt.step()
@staticmethod
def optimize(opt, loss, model=None, retain_graph=False):
opt.zero_grad()
loss.backward(retain_graph=retain_graph)
if model is not None:
torch.nn.utils.clip_grad_norm_(model.parameters(), cfg.clip_norm)
opt.step()
def show_config(self):
self.log.info(100 * '=')
self.log.info('> training arguments:')
for arg in vars(self.opt):
self.log.info('>>> {0}: {1}'.format(arg, getattr(self.opt, arg)))
self.log.info(100 * '=')
def cal_metrics(self, fmt_str=False):
"""
Calculate metrics
:param fmt_str: if return format string for logging
"""
with torch.no_grad():
# Prepare data for evaluation
eval_samples = self.gen.sample(cfg.samples_num, 4 * cfg.batch_size)
gen_data = GenDataIter(eval_samples)
gen_tokens = tensor_to_tokens(eval_samples, self.idx2word_dict)
gen_tokens_s = tensor_to_tokens(self.gen.sample(200, 200),
self.idx2word_dict)
# Reset metrics
self.bleu.reset(test_text=gen_tokens,
real_text=self.test_data.tokens)
self.nll_gen.reset(self.gen, self.train_data.loader)
self.nll_div.reset(self.gen, gen_data.loader)
self.self_bleu.reset(test_text=gen_tokens_s, real_text=gen_tokens)
self.ppl.reset(gen_tokens)
if fmt_str:
return ', '.join([
'%s = %s' % (metric.get_name(), metric.get_score())
for metric in self.all_metrics
])
else:
return [metric.get_score() for metric in self.all_metrics]
def cal_metrics_with_label(self, label_i):
assert type(label_i) == int, 'missing label'
with torch.no_grad():
# Prepare data for evaluation
eval_samples = self.gen.sample(cfg.samples_num,
8 * cfg.batch_size,
label_i=label_i)
gen_data = GenDataIter(eval_samples)
gen_tokens = tensor_to_tokens(eval_samples, self.idx2word_dict)
gen_tokens_s = tensor_to_tokens(
self.gen.sample(200, 200, label_i=label_i), self.idx2word_dict)
clas_data = CatClasDataIter([eval_samples], label_i)
# Reset metrics
self.bleu.reset(test_text=gen_tokens,
real_text=self.test_data_list[label_i].tokens)
self.nll_gen.reset(self.gen, self.train_data_list[label_i].loader,
label_i)
self.nll_div.reset(self.gen, gen_data.loader, label_i)
self.self_bleu.reset(test_text=gen_tokens_s, real_text=gen_tokens)
self.clas_acc.reset(self.clas, clas_data.loader)
self.ppl.reset(gen_tokens)
return [metric.get_score() for metric in self.all_metrics]
def comb_metrics(self, fmt_str=False):
all_scores = [
self.cal_metrics_with_label(label_i)
for label_i in range(cfg.k_label)
]
all_scores = np.array(
all_scores).T.tolist() # each row for each metric
if fmt_str:
return ', '.join([
'%s = %s' % (metric.get_name(), score)
for (metric, score) in zip(self.all_metrics, all_scores)
])
return all_scores
def _save(self, phase, epoch):
"""Save model state dict and generator's samples"""
if phase != 'ADV':
torch.save(
self.gen.state_dict(),
cfg.save_model_root + 'gen_{}_{:05d}.pt'.format(phase, epoch))
save_sample_path = cfg.save_samples_root + 'samples_{}_{}_{:05d}.txt'.format(
phase, cfg.samples_num, epoch)
samples = self.gen.sample(5000, cfg.batch_size)
write_tokens(save_sample_path,
tensor_to_tokens(samples, self.idx2word_dict))
def update_temperature(self, i, N):
self.gen.temperature.data = torch.Tensor(
[get_fixed_temperature(cfg.temperature, i, N, cfg.temp_adpt)])
if cfg.CUDA:
self.gen.temperature.data = self.gen.temperature.data.cuda()
def evaluation(self, eval_type):
"""Evaluation all children, update child score. Note that the eval data should be the same"""
eval_samples = [
self.gen.sample(cfg.eval_b_num * cfg.batch_size,
cfg.max_bn * cfg.batch_size,
label_i=i) for i in range(cfg.k_label)
]
# Fd
if cfg.lambda_fd != 0:
nll_div = []
for label_i in range(cfg.k_label):
gen_data = GenDataIter(eval_samples[label_i])
nll_div.append(
NLL.cal_nll_with_label(self.gen, gen_data.loader, label_i,
self.mle_criterion))
if 'f1' in eval_type:
if cfg.k_label == 1:
Fd = nll_div[0] if len(nll_div) > 0 else 0
elif cfg.k_label == 2:
Fd = nll_div[0] * nll_div[1] / (
nll_div[0] + nll_div[1]) if len(nll_div) > 0 else 0
else:
raise NotImplementedError("k_label = %d is not supported" %
cfg.k_label)
else:
Fd = sum(nll_div)
else:
Fd = 0
# Fq
if 'nll' in eval_type:
nll_oracle = []
for label_i in range(cfg.k_label):
gen_data = GenDataIter(eval_samples[label_i])
if cfg.lambda_fq != 0:
nll_oracle.append(-NLL.cal_nll_with_label(
self.oracle_list[label_i], gen_data.loader, label_i,
self.mle_criterion))
if 'f1' in eval_type:
if cfg.k_label == 1:
Fq = nll_oracle[0] if len(nll_oracle) > 0 else 0
elif cfg.k_label == 2:
Fq = nll_oracle[0] * nll_oracle[1] / (
nll_oracle[0] +
nll_oracle[1]) if len(nll_oracle) > 0 else 0
else:
raise NotImplementedError("k_label = %d is not supported" %
cfg.k_label)
else: # sum
Fq = sum(nll_oracle)
elif eval_type == 'Ra':
g_loss = 0
for i in range(cfg.k_label):
g_loss += torch.sigmoid(
self.eval_d_out_fake[i] -
torch.mean(self.eval_d_out_real[i])).sum()
Fq = g_loss.item()
else:
raise NotImplementedError("Evaluation '%s' is not implemented" %
eval_type)
score = cfg.lambda_fq * Fq + cfg.lambda_fd * Fd
return Fq, Fd, score