def __init__(self, model, sv, tv, optim, trg_dict_size,
valid_data=None, tests_data=None, n_critic=1):
self.lamda = 5
self.eps = 1e-20
#self.beta_KL = 0.005
self.beta_KL = 0.
self.beta_RLGen = 0.2
self.clip_rate = 0.
self.beta_RLBatch = 0.
self.model = model
self.decoder = model.decoder
self.classifier = self.decoder.classifier
self.sv, self.tv = sv, tv
self.trg_dict_size = trg_dict_size
self.n_critic = 1
self.translator_sample = Translator(self.model, sv, tv, k=1, noise=False)
#self.translator = Translator(model, sv, tv, k=10)
if isinstance(optim, list):
self.optim_G, self.optim_D = optim[0], optim[1]
self.optim_G.init_optimizer(self.model.parameters())
self.optim_D.init_optimizer(self.model.parameters())
else:
self.optim_G = Optim(
'adam', 10e-05, wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu
)
self.optim_G.init_optimizer(self.model.parameters())
self.optim_D = optim
self.optim_D.init_optimizer(self.model.parameters())
self.optim = [self.optim_G, self.optim_D]
'''
self.optim_RL = Optim(
'adadelta', 1.0, wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu
)
self.optim_RL.init_optimizer(self.model.parameters())
'''
self.maskSoftmax = MaskSoftmax()
self.valid_data = valid_data
self.tests_data = tests_data
class Trainer:
def __init__(self, model, sv, tv, optim, trg_dict_size,
valid_data=None, tests_data=None, n_critic=1):
self.lamda = 5
self.eps = 1e-20
#self.beta_KL = 0.005
self.beta_KL = 0.
self.beta_RLGen = 0.2
self.clip_rate = 0.
self.beta_RLBatch = 0.
self.model = model
self.decoder = model.decoder
self.classifier = self.decoder.classifier
self.sv, self.tv = sv, tv
self.trg_dict_size = trg_dict_size
self.n_critic = 1
self.translator_sample = Translator(self.model, sv, tv, k=1, noise=False)
#self.translator = Translator(model, sv, tv, k=10)
if isinstance(optim, list):
self.optim_G, self.optim_D = optim[0], optim[1]
self.optim_G.init_optimizer(self.model.parameters())
self.optim_D.init_optimizer(self.model.parameters())
else:
self.optim_G = Optim(
'adam', 10e-05, wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu
)
self.optim_G.init_optimizer(self.model.parameters())
self.optim_D = optim
self.optim_D.init_optimizer(self.model.parameters())
self.optim = [self.optim_G, self.optim_D]
'''
self.optim_RL = Optim(
'adadelta', 1.0, wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu
)
self.optim_RL.init_optimizer(self.model.parameters())
'''
self.maskSoftmax = MaskSoftmax()
self.valid_data = valid_data
self.tests_data = tests_data
def mt_eval(self, eid, bid, optim=None):
if optim: self.optim = optim
state_dict = { 'model': self.model.state_dict(), 'epoch': eid, 'batch': bid, 'optim': self.optim }
if wargs.save_one_model: model_file = '{}.pt'.format(wargs.model_prefix)
else: model_file = '{}_e{}_upd{}.pt'.format(wargs.model_prefix, eid, bid)
tc.save(state_dict, model_file)
wlog('Saving temporary model in {}'.format(model_file))
self.model.eval()
tor0 = Translator(self.model, self.sv, self.tv, print_att=wargs.print_att)
BLEU = tor0.trans_eval(self.valid_data, eid, bid, model_file, self.tests_data)
self.model.train()
return BLEU
# p1: (max_tlen_batch, batch_size, vocab_size)
def distance(self, P, Q, y_masks, type='JS', y_gold=None):
B = y_masks.size(1)
hypo_N = y_masks.data.sum()
if Q.size(0) > P.size(0): Q = Q[:(P.size(0) + 1)]
if type == 'JS':
#D_kl = tc.mean(tc.sum((tc.log(p1) - tc.log(p2)) * p1, dim=-1).squeeze(), dim=0)
M = (P + Q) / 2.
D_kl1 = tc.sum((tc.log(P) - tc.log(M)) * P, dim=-1).squeeze()
D_kl2 = tc.sum((tc.log(Q) - tc.log(M)) * Q, dim=-1).squeeze()
Js = 0.5 * D_kl1 + 0.5 * D_kl2
sent_batch_dist = tc.sum(Js * y_masks) / B
Js = Js / y_masks.sum(0)[None, :]
word_level_dist = tc.sum(Js * y_masks) / B
del M, D_kl1, D_kl2, Js
elif type == 'KL':
KL = tc.sum(P * (tc.log(P + self.eps) - tc.log(Q + self.eps)), dim=-1)
# (L, B, V) -> (L, B)
sent_batch_dist = tc.sum(KL * y_masks) / B
word_level_dist0 = tc.sum(KL * y_masks) / hypo_N
KL = KL / y_masks.sum(0)[None, :]
#print W_KL.data
word_level_dist1 = tc.sum(KL * y_masks) / B
#print W_dist.data[0], y_masks.size(1)
del KL
elif type == 'KL-sent':
#print p1[0]
#print p2[0]
#print '-----------------------------'
p1 = tc.gather(p1, 2, y_gold[:, :, None])[:, :, 0]
p2 = tc.gather(p2, 2, y_gold[:, :, None])[:, :, 0]
# p1 (max_tlen_batch, batch_size)
#print (p2 < 1) == False
KL = (y_masks * (tc.log(p1) - tc.log(p2))) * p1
sent_batch_dist = tc.sum(KL) / B
KL = KL / y_masks.sum(0)[None, :]
word_level_dist = tc.sum(KL * y_masks) / B
# KL: (1, batch_size)
del p1, p2, KL
return sent_batch_dist, word_level_dist0, word_level_dist1
def hyps_padding_dist(self, oracle, hyps_L, y_gold_maxL, p_y_hyp):
#hyps_dist = [None] * B
B, hyps_dist, hyps = oracle.size(1), [], [] # oracle, w/o bos
assert (B == len(hyps_L)) and (oracle.size(0) == p_y_hyp.size(0))
for bidx in range(B):
hyp_L = hyps_L[bidx] - 1 # remove bos
if hyp_L < y_gold_maxL:
padding = tc.ones(y_gold_maxL - hyp_L) / self.trg_dict_size
padding = padding[:, None].expand(padding.size(0), self.trg_dict_size)
#pad = pad[:, None].expand((pad.size(0), one_p_y_hyp.size(-1)))
padding = Variable(padding, requires_grad=False)
if wargs.gpu_id and not padding.is_cuda: padding = padding.cuda()
#print one_p_y_hyp.size(0), pad.size(0)
#print tc.cat((p_y_hyp[:hyp_L, bidx, :], padding), dim=0).size()
hyps_dist.append(tc.cat((p_y_hyp[:hyp_L, bidx, :], padding), dim=0))
hyps.append(tc.cat((oracle[:hyp_L, bidx],
Variable(PAD * tc.ones(y_gold_maxL - hyp_L).long()).cuda()), dim=0))
else:
hyps_dist.append(p_y_hyp[:y_gold_maxL, bidx, :])
hyps.append(oracle[:y_gold_maxL, bidx])
#hyps_dist[bidx] = one_p_y_hyp
hyps_dist = tc.stack(hyps_dist, dim=1)
hyps = tc.stack(hyps, dim=1)
return hyps_dist, hyps
def gumbel_sampling(self, B, y_maxL, feed_gold_out, noise=False):
# feed_gold_out (L * B, V)
logit = self.classifier.pred_map(feed_gold_out, noise=noise)
if logit.is_cuda: logit = logit.cpu()
hyps = tc.max(logit, 1)[1]
# hyps (L*B, 1)
hyps = hyps.view(y_maxL, B)
hyps[0] = BOS * tc.ones(B).long() # first words are <s>
# hyps (L, B)
c1 = tc.clamp((hyps.data - EOS), min=0, max=self.trg_dict_size)
c2 = tc.clamp((EOS - hyps.data), min=0, max=self.trg_dict_size)
_hyps = c1 + c2
_hyps = tc.cat([_hyps, tc.zeros(B).long().unsqueeze(0)], 0)
_hyps = tc.min(_hyps, 0)[1]
#_hyps = tc.max(0 - _hyps, 0)[1]
# idx: (1, B)
hyps_L = _hyps.view(-1).tolist()
hyps_mask = tc.zeros(y_maxL, B)
for bid in range(B): hyps_mask[:, bid][:hyps_L[bid]] = 1.
hyps_mask = Variable(hyps_mask, requires_grad=False)
if wargs.gpu_id and not hyps_mask.is_cuda: hyps_mask = hyps_mask.cuda()
if wargs.gpu_id and not hyps.is_cuda: hyps = hyps.cuda()
return hyps, hyps_mask, hyps_L
def try_trans(self, srcs, ref):
# (len, 1)
#src = sent_filter(list(srcs[:, bid].data))
x_filter = sent_filter(list(srcs))
y_filter = sent_filter(list(ref))
#wlog('\n[{:3}] {}'.format('Src', idx2sent(x_filter, self.sv)))
#wlog('[{:3}] {}'.format('Ref', idx2sent(y_filter, self.tv)))
onebest, onebest_ids, _ = self.translator_sample.trans_onesent(x_filter)
#wlog('[{:3}] {}'.format('Out', onebest))
# no EOS and BOS
return onebest_ids
def beamsearch_sampling(self, srcs, trgs, eos=True):
# y_masks: (trg_max_len, batch_size)
B = srcs.size(1)
oracles, oracles_L = [None] * B, [None] * B
for bidx in range(B):
onebest_ids = self.try_trans(srcs[:, bidx].data, trgs[:, bidx].data)
if len(onebest_ids) == 0 or onebest_ids[0] != BOS: onebest_ids = [BOS] + onebest_ids
if eos is True and onebest_ids[-1] != EOS: onebest_ids = onebest_ids + [EOS]
oracles_L[bidx] = len(onebest_ids)
oracles[bidx] = onebest_ids
maxL = max(oracles_L)
for bidx in range(B):
cur_L, oracle = oracles_L[bidx], oracles[bidx]
if cur_L < maxL: oracles[bidx] = oracle + [PAD] * (maxL - cur_L)
oracles = Variable(tc.Tensor(oracles).long().t(), requires_grad=False) # -> (L, B)
if wargs.gpu_id and not oracles.is_cuda: oracles = oracles.cuda()
oracles_mask = oracles.ne(PAD).float()
return oracles, oracles_mask, oracles_L
def train(self, dh, dev_input, k, merge=False, name='default', percentage=0.1):
#if (k + 1) % 1 == 0 and self.valid_data and self.tests_data:
# wlog('Evaluation on dev ... ')
# mt_eval(valid_data, self.model, self.sv, self.tv,
# 0, 0, [self.optim, self.optim_RL, self.optim_G], self.tests_data)
batch_count = len(dev_input)
self.model.train()
self.sampler = Nbs(self.model, self.tv, k=3, noise=False, print_att=False, batch_sample=True)
for eid in range(wargs.start_epoch, wargs.max_epochs + 1):
#self.optim_G.init_optimizer(self.model.parameters())
#self.optim_RL.init_optimizer(self.model.parameters())
size = int(percentage * batch_count)
shuffled_batch_idx = tc.randperm(batch_count)
wlog('{} NEW Epoch {}'.format('-' * 50, '-' * 50))
wlog('{}, Epo:{:>2}/{:>2} start, random {}/{}({:.2%}) calc BLEU ... '.format(
name, eid, wargs.max_epochs, size, batch_count, percentage), False)
param_1, param_2, param_3, param_4, param_5, param_6 = [], [], [], [], [], []
for k in range(size):
bid, half_size = shuffled_batch_idx[k], wargs.batch_size
# srcs: (max_sLen_batch, batch_size, emb), trgs: (max_tLen_batch, batch_size, emb)
if merge is False: _, srcs, _, trgs, _, slens, srcs_m, trgs_m = dev_input[bid]
else: _, srcs, _, trgs, _, slens, srcs_m, trgs_m = dh.merge_batch(dev_input[bid])[0]
trgs, trgs_m = trgs[0], trgs_m[0] # we only use the first dev reference
if wargs.sampling == 'gumbeling':
oracles, oracles_mask, oracles_L = self.gumbel_sampling(B, y_gold_maxL, feed_gold_out, True)
elif wargs.sampling == 'truncation':
oracles, oracles_mask, oracles_L = self.beamsearch_sampling(srcs, trgs)
elif wargs.sampling == 'length_limit':
batch_beam_trgs = self.sampler.beam_search_trans(srcs, srcs_m, trgs_m)
hyps = [list(zip(*b)[0]) for b in batch_beam_trgs]
oracles = batch_search_oracle(hyps, trgs[1:], trgs_m[1:])
if wargs.gpu_id and not oracles.is_cuda: oracles = oracles.cuda()
oracles_mask = oracles.ne(0).float()
oracles_L = oracles_mask.sum(0).data.int().tolist()
# oracles same with trgs, with bos and eos,(L, B)
param_1.append(BLToStrList(oracles[1:-1].t(), [l-2 for l in oracles_L]))
param_2.append(BLToStrList(trgs[1:-1].t(), trgs_m[1:-1].sum(0).data.int().tolist()))
param_3.append(BLToStrList(oracles[1:-1, :half_size].t(),
[l-2 for l in oracles_L[:half_size]]))
param_4.append(BLToStrList(trgs[1:-1, :half_size].t(),
trgs_m[1:-1, :half_size].sum(0).data.int().tolist()))
param_5.append(BLToStrList(oracles[1:-1, half_size:].t(),
[l-2 for l in oracles_L[half_size:]]))
param_6.append(BLToStrList(trgs[1:-1, half_size:].t(),
trgs_m[1:-1, half_size:].sum(0).data.int().tolist()))
start_bat_bleu_hist = bleu('\n'.join(param_3), ['\n'.join(param_4)], logfun=debug)
start_bat_bleu_new = bleu('\n'.join(param_5), ['\n'.join(param_6)], logfun=debug)
start_bat_bleu = bleu('\n'.join(param_1), ['\n'.join(param_2)], logfun=debug)
wlog('Random BLEU on history {}, new {}, mix {}'.format(
start_bat_bleu_hist, start_bat_bleu_new, start_bat_bleu))
wlog('Model selection and testing ... ')
self.mt_eval(eid, 0, [self.optim_G, self.optim_D])
if start_bat_bleu > 0.9:
wlog('Better BLEU ... go to next data history ...')
return
s_kl_seen, w_kl_seen0, w_kl_seen1, rl_gen_seen, rl_rho_seen, rl_bat_seen, w_mle_seen, \
s_mle_seen, ppl_seen = 0., 0., 0., 0., 0., 0., 0., 0., 0.
for bid in range(batch_count):
if merge is False: _, srcs, _, trgs, _, slens, srcs_m, trgs_m = dev_input[bid]
else: _, srcs, _, trgs, _, slens, srcs_m, trgs_m = dh.merge_batch(dev_input[bid], True)[0]
trgs, trgs_m = trgs[0], trgs_m[0]
gold_feed, gold_feed_mask = trgs[:-1], trgs_m[:-1]
gold, gold_mask = trgs[1:], trgs_m[1:]
B, y_gold_maxL = srcs.size(1), gold_feed.size(0)
N = gold.data.ne(PAD).sum()
debug('B:{}, gold_feed_ymaxL:{}, N:{}'.format(B, y_gold_maxL, N))
###################################################################################
debug('Optimizing KL distance ................................ {}'.format(name))
#self.model.zero_grad()
self.optim_G.zero_grad()
feed_gold_out, _ = self.model(srcs, gold_feed, srcs_m, gold_feed_mask)
p_y_gold = self.classifier.logit_to_prob(feed_gold_out)
# p_y_gold: (gold_max_len - 1, B, trg_dict_size)
if wargs.sampling == 'gumbeling':
oracles, oracles_mask, oracles_L = self.gumbel_sampling(B, y_gold_maxL, feed_gold_out, True)
elif wargs.sampling == 'truncation':
oracles, oracles_mask, oracles_L = self.beamsearch_sampling(srcs, trgs)
elif wargs.sampling == 'length_limit':
# w/o eos
batch_beam_trgs = self.sampler.beam_search_trans(srcs, srcs_m, trgs_m)
hyps = [list(zip(*b)[0]) for b in batch_beam_trgs]
oracles = batch_search_oracle(hyps, trgs[1:], trgs_m[1:])
if wargs.gpu_id and not oracles.is_cuda: oracles = oracles.cuda()
oracles_mask = oracles.ne(0).float()
oracles_L = oracles_mask.sum(0).data.int().tolist()
oracle_feed, oracle_feed_mask = oracles[:-1], oracles_mask[:-1]
oracle, oracle_mask = oracles[1:], oracles_mask[1:]
# oracles same with trgs, with bos and eos,(L, B)
feed_oracle_out, _ = self.model(srcs, oracle_feed, srcs_m, oracle_feed_mask)
p_y_hyp = self.classifier.logit_to_prob(feed_oracle_out)
p_y_hyp_pad, oracle = self.hyps_padding_dist(oracle, oracles_L, y_gold_maxL, p_y_hyp)
#wlog('feed oracle dist: {}, feed gold dist: {}, oracle: {}'.format(p_y_hyp_pad.size(), p_y_gold.size(), oracle.size()))
#B_KL_loss = self.distance(p_y_gold, p_y_hyp_pad, hyps_mask[1:], type='KL', y_gold=gold)
S_KL_loss, W_KL_loss0, W_KL_loss1 = self.distance(
p_y_gold, p_y_hyp_pad, gold_mask, type='KL', y_gold=gold)
debug('KL: Sent-level {}, Word0-level {}, Word1-level {}'.format(
S_KL_loss.data[0], W_KL_loss0.data[0], W_KL_loss1.data[0]))
s_kl_seen += S_KL_loss.data[0]
w_kl_seen0 += W_KL_loss0.data[0]
w_kl_seen1 += W_KL_loss1.data[0]
del p_y_hyp, feed_oracle_out
###################################################################################
debug('Optimizing RL(Gen) .......... {}'.format(name))
hyps_list = BLToStrList(oracle[:-1].t(), [l-2 for l in oracles_L], True)
trgs_list = BLToStrList(trgs[1:-1].t(), trgs_m[1:-1].sum(0).data.int().tolist(), True)
bleus_sampling = []
for hyp, ref in zip(hyps_list, trgs_list):
bleus_sampling.append(bleu(hyp, [ref], logfun=debug))
bleus_sampling = toVar(bleus_sampling, wargs.gpu_id)
oracle_mask = oracle.ne(0).float()
p_y_ahyp = p_y_hyp_pad.gather(2, oracle[:, :, None])[:, :, 0]
p_y_ahyp = ((p_y_ahyp + self.eps).log() * oracle_mask).sum(0) / oracle_mask.sum(0)
p_y_agold = p_y_gold.gather(2, gold[:, :, None])[:, :, 0]
p_y_agold = ((p_y_agold + self.eps).log() * gold_mask).sum(0) / gold_mask.sum(0)
r_theta = p_y_ahyp / p_y_agold
A = 1. - bleus_sampling
RL_Gen_loss = tc.min(r_theta * A, clip(r_theta, self.clip_rate) * A).sum()
RL_Gen_loss = (RL_Gen_loss).div(B)
debug('...... RL(Gen) cliped loss {}'.format(RL_Gen_loss.data[0]))
rl_gen_seen += RL_Gen_loss.data[0]
del p_y_agold
###################################################################################
debug('Optimizing RL(Batch) -> Gap of MLE and BLEU ... rho ... feed onebest .... ')
param_1 = BLToStrList(oracles[1:-1].t(), [l-2 for l in oracles_L])
param_2 = BLToStrList(trgs[1:-1].t(), trgs_m[1:-1].sum(0).data.int().tolist())
rl_bat_bleu = bleu(param_1, [param_2], logfun=debug)
rl_avg_bleu = tc.mean(bleus_sampling).data[0]
rl_rho = cor_coef(p_y_ahyp, bleus_sampling, eps=self.eps)
rl_rho_seen += rl_rho.data[0] # must use data, accumulating Variable needs more memory
#p_y_hyp = p_y_hyp.exp()
#p_y_hyp = (p_y_hyp * self.lamda / 3).exp()
#p_y_hyp = self.maskSoftmax(p_y_hyp)
p_y_ahyp = p_y_ahyp[None, :]
p_y_ahyp_T = p_y_ahyp.t().expand(B, B)
p_y_ahyp = p_y_ahyp.expand(B, B)
p_y_ahyp_sum = p_y_ahyp_T + p_y_ahyp + self.eps
#bleus_sampling = bleus_sampling[None, :].exp()
bleus_sampling = self.maskSoftmax(self.lamda * bleus_sampling[None, :])
bleus_T = bleus_sampling.t().expand(B, B)
bleus = bleus_sampling.expand(B, B)
bleus_sum = bleus_T + bleus + self.eps
#print 'p_y_hyp_sum......................'
#print p_y_hyp_sum.data
RL_Batch_loss = p_y_ahyp / p_y_ahyp_sum * tc.log(bleus_T / bleus_sum) + \
p_y_ahyp_T / p_y_ahyp_sum * tc.log(bleus / bleus_sum)
#RL_Batch_loss = tc.sum(-RL_Batch_loss * toVar(1 - tc.eye(B))).div(B)
RL_Batch_loss = tc.sum(-RL_Batch_loss * toVar(1 - tc.eye(B), wargs.gpu_id))
debug('RL(Batch) Mean BLEU: {}, rl_batch_loss: {}, rl_rho: {}, Bat BLEU: {}'.format(
rl_avg_bleu, RL_Batch_loss.data[0], rl_rho.data[0], rl_bat_bleu))
rl_bat_seen += RL_Batch_loss.data[0]
del oracles, oracles_mask, oracle_feed, oracle_feed_mask, oracle, oracle_mask,\
p_y_ahyp, bleus_sampling, bleus, p_y_ahyp_T, p_y_ahyp_sum, bleus_T, bleus_sum
'''
(self.beta_KL * S_KL_loss + self.beta_RLGen * RL_Gen_loss + \
self.beta_RLBatch * RL_Batch_loss).backward(retain_graph=True)
mle_loss, grad_output, _ = memory_efficient(
feed_gold_out, gold, gold_mask, self.model.classifier)
feed_gold_out.backward(grad_output)
'''
(self.beta_KL * W_KL_loss0 + self.beta_RLGen * RL_Gen_loss + \
self.beta_RLBatch * RL_Batch_loss).backward(retain_graph=True)
self.optim_G.step()
###################################################### discrimitor
#mle_loss, _, _ = self.classifier(feed_gold_out, gold, gold_mask)
#mle_loss = mle_loss.div(B)
#mle_loss = mle_loss.data[0]
self.optim_D.zero_grad()
mle_loss, _, _ = self.classifier.snip_back_prop(feed_gold_out, gold, gold_mask)
self.optim_D.step()
w_mle_seen += ( mle_loss / N )
s_mle_seen += ( mle_loss / B )
ppl_seen += math.exp(mle_loss/N)
wlog('Epo:{:>2}/{:>2}, Bat:[{}/{}], W0-KL {:4.2f}, W1-KL {:4.2f}, '
'S-RLGen {:4.2f}, B-rho {:4.2f}, B-RLBat {:4.2f}, W-MLE:{:4.2f}, '
'S-MLE:{:4.2f}, W-ppl:{:4.2f}, B-bleu:{:4.2f}, A-bleu:{:4.2f}'.format(
eid, wargs.max_epochs, bid, batch_count, W_KL_loss0.data[0],
W_KL_loss1.data[0], RL_Gen_loss.data[0], rl_rho.data[0], RL_Batch_loss.data[0],
mle_loss/N, mle_loss/B, math.exp(mle_loss/N), rl_bat_bleu, rl_avg_bleu))
#wlog('=' * 100)
del S_KL_loss, W_KL_loss0, W_KL_loss1, RL_Gen_loss, RL_Batch_loss, feed_gold_out
wlog('End epoch: S-KL {:4.2f}, W0-KL {:4.2f}, W1-KL {:4.2f}, S-RLGen {:4.2f}, B-rho '
'{:4.2f}, B-RLBat {:4.2f}, W-MLE {:4.2f}, S-MLE {:4.2f}, W-ppl {:4.2f}'.format(
s_kl_seen/batch_count, w_kl_seen0/batch_count, w_kl_seen1/batch_count, rl_gen_seen/batch_count,
rl_rho_seen/batch_count, rl_bat_seen/batch_count, w_mle_seen/batch_count,
s_mle_seen/batch_count, ppl_seen/batch_count))
def main():
init_dir(wargs.dir_model)
init_dir(wargs.dir_valid)
vocab_data = {}
train_srcD_file = wargs.src_vocab_from
wlog('\nPreparing out of domain source vocabulary from {} ... '.format(
train_srcD_file))
src_vocab = extract_vocab(train_srcD_file, wargs.src_dict,
wargs.src_dict_size)
#DANN
train_srcD_file_domain = wargs.src_domain_vocab_from
wlog('\nPreparing in domain source vocabulary from {} ...'.format(
train_srcD_file_domain))
src_vocab = updata_vocab(train_srcD_file_domain, src_vocab, wargs.src_dict,
wargs.src_dict_size)
vocab_data['src'] = src_vocab
train_trgD_file = wargs.trg_vocab_from
wlog('\nPreparing out of domain target vocabulary from {} ... '.format(
train_trgD_file))
trg_vocab = extract_vocab(train_trgD_file, wargs.trg_dict,
wargs.trg_dict_size)
#DANN
train_trgD_file_domain = wargs.trg_domain_vocab_from
wlog('\nPreparing in domain target vocabulary from {} ... '.format(
train_trgD_file_domain))
trg_vocab = updata_vocab(train_trgD_file_domain, trg_vocab, wargs.trg_dict,
wargs.trg_dict_size)
vocab_data['trg'] = trg_vocab
train_src_file = wargs.train_src
train_trg_file = wargs.train_trg
if wargs.fine_tune is False:
wlog('\nPreparing out of domain training set from {} and {} ... '.
format(train_src_file, train_trg_file))
train_src_tlst, train_trg_tlst = wrap_data(
train_src_file,
train_trg_file,
vocab_data['src'],
vocab_data['trg'],
max_seq_len=wargs.max_seq_len)
else:
wlog('\nNo out of domain trainin set ...')
#DANN
train_src_file_domain = wargs.train_src_domain
train_trg_file_domain = wargs.train_trg_domain
wlog('\nPreparing in domain training set from {} and {}...'.format(
train_src_file_domain, train_trg_file_domain))
train_src_tlst_domain, train_trg_tlst_domain = wrap_data(
train_src_file_domain,
train_trg_file_domain,
vocab_data['src'],
vocab_data['trg'],
max_seq_len=wargs.max_seq_len)
'''
list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...]
no padding
'''
valid_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.val_prefix,
wargs.val_src_suffix)
wlog('\nPreparing validation set from {} ... '.format(valid_file))
valid_src_tlst, valid_src_lens = val_wrap_data(valid_file, src_vocab)
if wargs.fine_tune is False:
wlog('Out of domain Sentence-pairs count in training data: {}'.format(
len(train_src_tlst)))
wlog('In domain Sentence-pairs count in training data: {}'.format(
len(train_src_tlst_domain)))
src_vocab_size, trg_vocab_size = vocab_data['src'].size(
), vocab_data['trg'].size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(
src_vocab_size, trg_vocab_size))
if wargs.fine_tune is False:
batch_train = Input(train_src_tlst, train_trg_tlst, wargs.batch_size)
else:
batch_train = None
batch_valid = Input(valid_src_tlst, None, 1, volatile=True)
#DANN
batch_train_domain = Input(train_src_tlst_domain, train_trg_tlst_domain,
wargs.batch_size)
tests_data = None
if wargs.tests_prefix is not None:
init_dir(wargs.dir_tests)
tests_data = {}
for prefix in wargs.tests_prefix:
init_dir(wargs.dir_tests + '/' + prefix)
test_file = '{}{}.{}'.format(wargs.val_tst_dir, prefix,
wargs.val_src_suffix)
wlog('Preparing test set from {} ... '.format(test_file))
test_src_tlst, _ = val_wrap_data(test_file, src_vocab)
tests_data[prefix] = Input(test_src_tlst, None, 1, volatile=True)
sv = vocab_data['src'].idx2key
tv = vocab_data['trg'].idx2key
nmtModel = NMT(src_vocab_size, trg_vocab_size)
if wargs.pre_train is not None:
assert os.path.exists(wargs.pre_train), 'Requires pre-trained model'
_dict = _load_model(wargs.pre_train)
# initializing parameters of interactive attention model
class_dict = None
if len(_dict) == 4: model_dict, eid, bid, optim = _dict
elif len(_dict) == 5:
model_dict, class_dict, eid, bid, optim = _dict
for name, param in nmtModel.named_parameters():
if name in model_dict:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(model_dict[name])
wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad,
name))
elif name.endswith('map_vocab.weight'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.weight'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
elif name.endswith('map_vocab.bias'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.bias'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
else:
init_params(param, name, True)
wargs.start_epoch = eid + 1
else:
for n, p in nmtModel.named_parameters():
init_params(p, n, True)
optim = Optim(wargs.opt_mode,
wargs.learning_rate,
wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu)
if wargs.gpu_id:
nmtModel.cuda()
wlog('Push model onto GPU[{}] ... '.format(wargs.gpu_id[0]))
else:
nmtModel.cpu()
wlog('Push model onto CPU ... ')
wlog(nmtModel)
wlog(optim)
pcnt1 = len([p for p in nmtModel.parameters()])
pcnt2 = sum([p.nelement() for p in nmtModel.parameters()])
wlog('Parameters number: {}/{}'.format(pcnt1, pcnt2))
optim.init_optimizer(nmtModel.parameters())
trainer = Trainer(nmtModel, batch_train, batch_train_domain, vocab_data,
optim, batch_valid, tests_data)
trainer.train()
def main():
#if wargs.ss_type is not None: assert wargs.model == 1, 'Only rnnsearch support schedule sample'
init_dir(wargs.dir_model)
init_dir(wargs.dir_valid)
src = os.path.join(wargs.dir_data, '{}.{}'.format(wargs.train_prefix, wargs.train_src_suffix))
trg = os.path.join(wargs.dir_data, '{}.{}'.format(wargs.train_prefix, wargs.train_trg_suffix))
vocabs = {}
wlog('\nPreparing source vocabulary from {} ... '.format(src))
src_vocab = extract_vocab(src, wargs.src_vcb, wargs.n_src_vcb_plan,
wargs.max_seq_len, char=wargs.src_char)
wlog('\nPreparing target vocabulary from {} ... '.format(trg))
trg_vocab = extract_vocab(trg, wargs.trg_vcb, wargs.n_trg_vcb_plan, wargs.max_seq_len)
n_src_vcb, n_trg_vcb = src_vocab.size(), trg_vocab.size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(n_src_vcb, n_trg_vcb))
vocabs['src'], vocabs['trg'] = src_vocab, trg_vocab
wlog('\nPreparing training set from {} and {} ... '.format(src, trg))
trains = {}
train_src_tlst, train_trg_tlst = wrap_data(wargs.dir_data, wargs.train_prefix,
wargs.train_src_suffix, wargs.train_trg_suffix,
src_vocab, trg_vocab, shuffle=True,
sort_k_batches=wargs.sort_k_batches,
max_seq_len=wargs.max_seq_len,
char=wargs.src_char)
'''
list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...]
no padding
'''
batch_train = Input(train_src_tlst, train_trg_tlst, wargs.batch_size,
batch_type=wargs.batch_type, bow=wargs.trg_bow, batch_sort=False)
wlog('Sentence-pairs count in training data: {}'.format(len(train_src_tlst)))
batch_valid = None
if wargs.val_prefix is not None:
val_src_file = os.path.join(wargs.val_tst_dir, '{}.{}'.format(wargs.val_prefix, wargs.val_src_suffix))
val_trg_file = os.path.join(wargs.val_tst_dir, '{}.{}'.format(wargs.val_prefix, wargs.val_ref_suffix))
wlog('\nPreparing validation set from {} and {} ... '.format(val_src_file, val_trg_file))
valid_src_tlst, valid_trg_tlst = wrap_data(wargs.val_tst_dir, wargs.val_prefix,
wargs.val_src_suffix, wargs.val_ref_suffix,
src_vocab, trg_vocab, shuffle=False,
max_seq_len=wargs.dev_max_seq_len,
char=wargs.src_char)
batch_valid = Input(valid_src_tlst, valid_trg_tlst, 1, batch_sort=False)
batch_tests = None
if wargs.tests_prefix is not None:
assert isinstance(wargs.tests_prefix, list), 'Test files should be list.'
init_dir(wargs.dir_tests)
batch_tests = {}
for prefix in wargs.tests_prefix:
init_dir(wargs.dir_tests + '/' + prefix)
test_file = '{}{}.{}'.format(wargs.val_tst_dir, prefix, wargs.val_src_suffix)
wlog('\nPreparing test set from {} ... '.format(test_file))
test_src_tlst, _ = wrap_tst_data(test_file, src_vocab, char=wargs.src_char)
batch_tests[prefix] = Input(test_src_tlst, None, 1, batch_sort=False)
wlog('\n## Finish to Prepare Dataset ! ##\n')
src_emb = WordEmbedding(n_src_vcb, wargs.d_src_emb, wargs.input_dropout,
wargs.position_encoding, prefix='Src')
trg_emb = WordEmbedding(n_trg_vcb, wargs.d_trg_emb, wargs.input_dropout,
wargs.position_encoding, prefix='Trg')
# share the embedding matrix - preprocess with share_vocab required.
if wargs.embs_share_weight:
if n_src_vcb != n_trg_vcb:
raise AssertionError('The `-share_vocab` should be set during '
'preprocess if you use share_embeddings!')
src_emb.we.weight = trg_emb.we.weight
nmtModel = build_NMT(src_emb, trg_emb)
if not wargs.copy_attn:
classifier = Classifier(wargs.d_model if wargs.decoder_type == 'att' else 2 * wargs.d_enc_hid,
n_trg_vcb, trg_emb, loss_norm=wargs.loss_norm,
label_smoothing=wargs.label_smoothing,
emb_loss=wargs.emb_loss, bow_loss=wargs.bow_loss)
nmtModel.decoder.classifier = classifier
if wargs.gpu_id is not None:
wlog('push model onto GPU {} ... '.format(wargs.gpu_id), 0)
#nmtModel = nn.DataParallel(nmtModel, device_ids=wargs.gpu_id)
nmtModel.to(tc.device('cuda'))
else:
wlog('push model onto CPU ... ', 0)
nmtModel.to(tc.device('cpu'))
wlog('done.')
if wargs.pre_train is not None:
assert os.path.exists(wargs.pre_train)
from tools.utils import load_model
_dict = load_model(wargs.pre_train)
# initializing parameters of interactive attention model
class_dict = None
if len(_dict) == 5:
model_dict, class_dict, eid, bid, optim = _dict
elif len(_dict) == 4:
model_dict, eid, bid, optim = _dict
for name, param in nmtModel.named_parameters():
if name in model_dict:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(model_dict[name])
wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad, name))
elif name.endswith('map_vocab.weight'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.weight'])
wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad, name))
elif name.endswith('map_vocab.bias'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.bias'])
wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad, name))
else: init_params(param, name, init_D=wargs.param_init_D, a=float(wargs.u_gain))
wargs.start_epoch = eid + 1
else:
optim = Optim(wargs.opt_mode, wargs.learning_rate, wargs.max_grad_norm)
#for n, p in nmtModel.named_parameters():
# bias can not be initialized uniformly
#if wargs.encoder_type != 'att' and wargs.decoder_type != 'att':
# init_params(p, n, init_D=wargs.param_init_D, a=float(wargs.u_gain))
wlog(nmtModel)
wlog(optim)
pcnt1 = len([p for p in nmtModel.parameters()])
pcnt2 = sum([p.nelement() for p in nmtModel.parameters()])
wlog('parameters number: {}/{}'.format(pcnt1, pcnt2))
wlog('\n' + '*' * 30 + ' trainable parameters ' + '*' * 30)
for n, p in nmtModel.named_parameters():
if p.requires_grad: wlog('{:60} : {}'.format(n, p.size()))
optim.init_optimizer(nmtModel.parameters())
trainer = Trainer(nmtModel, batch_train, vocabs, optim, batch_valid, batch_tests)
trainer.train()
def main():
# if wargs.ss_type is not None: assert wargs.model == 1, 'Only rnnsearch support schedule sample'
init_dir(wargs.dir_model)
init_dir(wargs.dir_valid)
src = os.path.join(
wargs.dir_data, '{}.{}'.format(wargs.train_prefix,
wargs.train_src_suffix))
trg = os.path.join(
wargs.dir_data, '{}.{}'.format(wargs.train_prefix,
wargs.train_trg_suffix))
src, trg = os.path.abspath(src), os.path.abspath(trg)
vocabs = {}
if wargs.share_vocab is False:
wlog('\nPreparing source vocabulary from {} ... '.format(src))
src_vocab = extract_vocab(src,
wargs.src_vcb,
wargs.n_src_vcb_plan,
wargs.max_seq_len,
char=wargs.src_char)
wlog('\nPreparing target vocabulary from {} ... '.format(trg))
trg_vocab = extract_vocab(trg, wargs.trg_vcb, wargs.n_trg_vcb_plan,
wargs.max_seq_len)
n_src_vcb, n_trg_vcb = src_vocab.size(), trg_vocab.size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(
n_src_vcb, n_trg_vcb))
else:
wlog('\nPreparing the shared vocabulary from \n\t{}\n\t{}'.format(
src, trg))
trg_vocab = src_vocab = extract_vocab(src,
wargs.src_vcb,
wargs.n_src_vcb_plan,
wargs.max_seq_len,
share_vocab=True,
trg_file=trg)
n_src_vcb, n_trg_vcb = src_vocab.size(), trg_vocab.size()
wlog('Shared vocabulary size: |vocab|={}'.format(src_vocab.size()))
vocabs['src'], vocabs['trg'] = src_vocab, trg_vocab
wlog('\nPreparing training set from {} and {} ... '.format(src, trg))
trains = {}
train_src_tlst, train_trg_tlst = wrap_data(
wargs.dir_data,
wargs.train_prefix,
wargs.train_src_suffix,
wargs.train_trg_suffix,
src_vocab,
trg_vocab,
shuffle=True,
sort_k_batches=wargs.sort_k_batches,
max_seq_len=wargs.max_seq_len,
char=wargs.src_char)
'''
list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...]
no padding
'''
batch_train = Input(train_src_tlst,
train_trg_tlst,
wargs.batch_size,
batch_type=wargs.batch_type,
bow=wargs.trg_bow,
batch_sort=False,
gpu_ids=device_ids)
wlog('Sentence-pairs count in training data: {}'.format(
len(train_src_tlst)))
batch_valid = None
if wargs.val_prefix is not None:
val_src_file = os.path.join(
wargs.val_tst_dir, '{}.{}'.format(wargs.val_prefix,
wargs.val_src_suffix))
val_trg_file = os.path.join(
wargs.val_tst_dir, '{}.{}'.format(wargs.val_prefix,
wargs.val_ref_suffix))
val_src_file, val_trg_file = os.path.abspath(
val_src_file), os.path.abspath(val_trg_file)
wlog('\nPreparing validation set from {} and {} ... '.format(
val_src_file, val_trg_file))
valid_src_tlst, valid_trg_tlst = wrap_data(
wargs.val_tst_dir,
wargs.val_prefix,
wargs.val_src_suffix,
wargs.val_ref_suffix,
src_vocab,
trg_vocab,
shuffle=False,
max_seq_len=wargs.dev_max_seq_len,
char=wargs.src_char)
batch_valid = Input(valid_src_tlst,
valid_trg_tlst,
batch_size=wargs.valid_batch_size,
batch_sort=False,
gpu_ids=device_ids)
batch_tests = None
if wargs.tests_prefix is not None:
assert isinstance(wargs.tests_prefix,
list), 'Test files should be list.'
init_dir(wargs.dir_tests)
batch_tests = {}
for prefix in wargs.tests_prefix:
init_dir(wargs.dir_tests + '/' + prefix)
test_file = '{}{}.{}'.format(wargs.val_tst_dir, prefix,
wargs.val_src_suffix)
test_file = os.path.abspath(test_file)
wlog('\nPreparing test set from {} ... '.format(test_file))
test_src_tlst, _ = wrap_tst_data(test_file,
src_vocab,
char=wargs.src_char)
batch_tests[prefix] = Input(test_src_tlst,
None,
batch_size=wargs.test_batch_size,
batch_sort=False,
gpu_ids=device_ids)
wlog('\n## Finish to Prepare Dataset ! ##\n')
src_emb = WordEmbedding(n_src_vcb,
wargs.d_src_emb,
wargs.input_dropout,
wargs.position_encoding,
prefix='Src')
trg_emb = WordEmbedding(n_trg_vcb,
wargs.d_trg_emb,
wargs.input_dropout,
wargs.position_encoding,
prefix='Trg')
# share the embedding matrix between the source and target
if wargs.share_vocab is True: src_emb.we.weight = trg_emb.we.weight
nmtModel = build_NMT(src_emb, trg_emb)
if device_ids is not None:
wlog('push model onto GPU {} ... '.format(device_ids[0]), 0)
nmtModel_par = nn.DataParallel(nmtModel, device_ids=device_ids)
nmtModel_par.to(device)
else:
wlog('push model onto CPU ... ', 0)
nmtModel.to(tc.device('cpu'))
wlog('done.')
if wargs.pre_train is not None:
wlog(wargs.pre_train)
assert os.path.exists(wargs.pre_train)
from tools.utils import load_model
_dict = load_model(wargs.pre_train)
# initializing parameters of interactive attention model
class_dict = None
if len(_dict) == 5:
# model_dict, e_idx, e_bidx, n_steps, optim = _dict['model'], _dict['epoch'], _dict['batch'], _dict['steps'], _dict['optim']
model_dict, e_idx, e_bidx, n_steps, optim = _dict
elif len(_dict) == 4:
model_dict, e_idx, e_bidx, optim = _dict
for name, param in nmtModel.named_parameters():
if name in model_dict:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(model_dict[name])
# wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad, name))
elif name.endswith('map_vocab.weight'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.weight'])
# wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad, name))
elif name.endswith('map_vocab.bias'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.bias'])
# wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad, name))
else:
init_params(param,
name,
init_D=wargs.param_init_D,
a=float(wargs.u_gain))
# wargs.start_epoch = e_idx + 1
# # 不重新开始
# optim.n_current_steps = 0
else:
optim = Optim(wargs.opt_mode, wargs.learning_rate, wargs.max_grad_norm)
for n, p in nmtModel.named_parameters():
# bias can not be initialized uniformly
if 'norm' in n:
wlog('ignore layer norm init ...')
continue
if 'emb' in n:
wlog('ignore word embedding weight init ...')
continue
if 'vcb_proj' in n:
wlog('ignore vcb_proj weight init ...')
continue
init_params(p, n, init_D=wargs.param_init_D, a=float(wargs.u_gain))
# if wargs.encoder_type != 'att' and wargs.decoder_type != 'att':
# init_params(p, n, init_D=wargs.param_init_D, a=float(wargs.u_gain))
# wlog(nmtModel)
wlog(optim)
pcnt1 = len([p for p in nmtModel.parameters()])
pcnt2 = sum([p.nelement() for p in nmtModel.parameters()])
wlog('parameters number: {}/{}'.format(pcnt1, pcnt2))
# wlog('\n' + '*' * 30 + ' trainable parameters ' + '*' * 30)
# for n, p in nmtModel.named_parameters():
# if p.requires_grad: wlog('{:60} : {}'.format(n, p.size()))
opt_state = None
if wargs.pre_train:
opt_state = optim.optimizer.state_dict()
if wargs.use_reinfore_ce is False:
criterion = LabelSmoothingCriterion(
trg_emb.n_vocab, label_smoothing=wargs.label_smoothing)
else:
word2vec = tc.load(wargs.word2vec_weight)['w2v']
# criterion = Word2VecDistanceCriterion(word2vec)
criterion = CosineDistance(word2vec)
if device_ids is not None:
wlog('push criterion onto GPU {} ... '.format(device_ids[0]), 0)
criterion = criterion.to(device)
wlog('done.')
# if wargs.reinfore_type == 0 or wargs.reinfore_type == 1:
# param = list(nmtModel.parameters())
# else:
# param = list(nmtModel.parameters()) + list(criterion.parameters())
param = list(nmtModel.parameters())
optim.init_optimizer(param)
lossCompute = MultiGPULossCompute(
nmtModel.generator,
criterion,
wargs.d_model if wargs.decoder_type == 'att' else 2 * wargs.d_enc_hid,
n_trg_vcb,
trg_emb,
nmtModel.bowMapper,
loss_norm=wargs.loss_norm,
chunk_size=wargs.chunk_size,
device_ids=device_ids)
trainer = Trainer(nmtModel_par, batch_train, vocabs, optim, lossCompute,
nmtModel, batch_valid, batch_tests, writer)
trainer.train()
writer.close()
def main():
#if wargs.ss_type is not None: assert wargs.model == 1, 'Only rnnsearch support schedule sample'
init_dir(wargs.dir_model)
init_dir(wargs.dir_valid)
src = os.path.join(
wargs.dir_data, '{}.{}'.format(wargs.train_prefix,
wargs.train_src_suffix))
trg = os.path.join(
wargs.dir_data, '{}.{}'.format(wargs.train_prefix,
wargs.train_trg_suffix))
vocabs = {}
wlog('\n[o/Subword] Preparing source vocabulary from {} ... '.format(src))
src_vocab = extract_vocab(src,
wargs.src_dict,
wargs.src_dict_size,
wargs.max_seq_len,
char=wargs.src_char)
wlog('\n[o/Subword] Preparing target vocabulary from {} ... '.format(trg))
trg_vocab = extract_vocab(trg, wargs.trg_dict, wargs.trg_dict_size,
wargs.max_seq_len)
src_vocab_size, trg_vocab_size = src_vocab.size(), trg_vocab.size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(
src_vocab_size, trg_vocab_size))
vocabs['src'], vocabs['trg'] = src_vocab, trg_vocab
wlog('\nPreparing training set from {} and {} ... '.format(src, trg))
trains = {}
train_src_tlst, train_trg_tlst = wrap_data(wargs.dir_data,
wargs.train_prefix,
wargs.train_src_suffix,
wargs.train_trg_suffix,
src_vocab,
trg_vocab,
max_seq_len=wargs.max_seq_len,
char=wargs.src_char)
'''
list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...]
no padding
'''
batch_train = Input(train_src_tlst,
train_trg_tlst,
wargs.batch_size,
batch_sort=True)
wlog('Sentence-pairs count in training data: {}'.format(
len(train_src_tlst)))
batch_valid = None
if wargs.val_prefix is not None:
val_src_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.val_prefix,
wargs.val_src_suffix)
val_trg_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.val_prefix,
wargs.val_ref_suffix)
wlog('\nPreparing validation set from {} and {} ... '.format(
val_src_file, val_trg_file))
valid_src_tlst, valid_trg_tlst = wrap_data(
wargs.val_tst_dir,
wargs.val_prefix,
wargs.val_src_suffix,
wargs.val_ref_suffix,
src_vocab,
trg_vocab,
shuffle=False,
sort_data=False,
max_seq_len=wargs.dev_max_seq_len,
char=wargs.src_char)
batch_valid = Input(valid_src_tlst,
valid_trg_tlst,
1,
volatile=True,
batch_sort=False)
batch_tests = None
if wargs.tests_prefix is not None:
assert isinstance(wargs.tests_prefix,
list), 'Test files should be list.'
init_dir(wargs.dir_tests)
batch_tests = {}
for prefix in wargs.tests_prefix:
init_dir(wargs.dir_tests + '/' + prefix)
test_file = '{}{}.{}'.format(wargs.val_tst_dir, prefix,
wargs.val_src_suffix)
wlog('\nPreparing test set from {} ... '.format(test_file))
test_src_tlst, _ = wrap_tst_data(test_file,
src_vocab,
char=wargs.src_char)
batch_tests[prefix] = Input(test_src_tlst,
None,
1,
volatile=True,
batch_sort=False)
wlog('\n## Finish to Prepare Dataset ! ##\n')
nmtModel = NMT(src_vocab_size, trg_vocab_size)
if wargs.pre_train is not None:
assert os.path.exists(wargs.pre_train)
_dict = _load_model(wargs.pre_train)
# initializing parameters of interactive attention model
class_dict = None
if len(_dict) == 4: model_dict, eid, bid, optim = _dict
elif len(_dict) == 5:
model_dict, class_dict, eid, bid, optim = _dict
for name, param in nmtModel.named_parameters():
if name in model_dict:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(model_dict[name])
wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad,
name))
elif name.endswith('map_vocab.weight'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.weight'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
elif name.endswith('map_vocab.bias'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.bias'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
else:
init_params(param, name, True)
wargs.start_epoch = eid + 1
else:
for n, p in nmtModel.named_parameters():
init_params(p, n, True)
optim = Optim(wargs.opt_mode,
wargs.learning_rate,
wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu,
model=wargs.model)
if wargs.gpu_id is not None:
wlog('Push model onto GPU {} ... '.format(wargs.gpu_id), 0)
nmtModel.cuda()
else:
wlog('Push model onto CPU ... ', 0)
nmtModel.cpu()
wlog('done.')
wlog(nmtModel)
wlog(optim)
pcnt1 = len([p for p in nmtModel.parameters()])
pcnt2 = sum([p.nelement() for p in nmtModel.parameters()])
wlog('Parameters number: {}/{}'.format(pcnt1, pcnt2))
optim.init_optimizer(nmtModel.parameters())
trainer = Trainer(nmtModel, batch_train, vocabs, optim, batch_valid,
batch_tests)
trainer.train()
def main():
# Check if CUDA is available
if cuda.is_available():
wlog('CUDA is available, specify device by gpu_id argument (i.e. gpu_id=[3])')
else:
wlog('Warning: CUDA is not available, try CPU')
if wargs.gpu_id:
cuda.set_device(wargs.gpu_id[0])
wlog('Using GPU {}'.format(wargs.gpu_id[0]))
init_dir(wargs.dir_model)
init_dir(wargs.dir_valid)
init_dir(wargs.dir_tests)
for prefix in wargs.tests_prefix:
if not prefix == wargs.val_prefix: init_dir(wargs.dir_tests + '/' + prefix)
wlog('Preparing data ... ', 0)
train_srcD_file = wargs.dir_data + 'train.10k.zh5'
wlog('\nPreparing source vocabulary from {} ... '.format(train_srcD_file))
src_vocab = extract_vocab(train_srcD_file, wargs.src_dict, wargs.src_dict_size)
train_trgD_file = wargs.dir_data + 'train.10k.en5'
wlog('\nPreparing target vocabulary from {} ... '.format(train_trgD_file))
trg_vocab = extract_vocab(train_trgD_file, wargs.trg_dict, wargs.trg_dict_size)
train_src_file = wargs.dir_data + 'train.10k.zh0'
train_trg_file = wargs.dir_data + 'train.10k.en0'
wlog('\nPreparing training set from {} and {} ... '.format(train_src_file, train_trg_file))
train_src_tlst, train_trg_tlst = wrap_data(train_src_file, train_trg_file, src_vocab, trg_vocab)
#list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...], no padding
wlog('Sentence-pairs count in training data: {}'.format(len(train_src_tlst)))
src_vocab_size, trg_vocab_size = src_vocab.size(), trg_vocab.size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(src_vocab_size, trg_vocab_size))
batch_train = Input(train_src_tlst, train_trg_tlst, wargs.batch_size)
tests_data = None
if wargs.tests_prefix is not None:
tests_data = {}
for prefix in wargs.tests_prefix:
test_file = wargs.val_tst_dir + prefix + '.src'
test_src_tlst, _ = val_wrap_data(test_file, src_vocab)
# we select best model by nist03 testing data
if prefix == wargs.val_prefix:
wlog('\nPreparing model-select set from {} ... '.format(test_file))
batch_valid = Input(test_src_tlst, None, 1, volatile=True, prefix=prefix)
else:
wlog('\nPreparing test set from {} ... '.format(test_file))
tests_data[prefix] = Input(test_src_tlst, None, 1, volatile=True)
nmtModel = NMT()
classifier = Classifier(wargs.out_size, trg_vocab_size)
if wargs.pre_train:
model_dict, class_dict, eid, bid, optim = load_pytorch_model(wargs.pre_train)
if isinstance(optim, list): _, _, optim = optim
# initializing parameters of interactive attention model
for p in nmtModel.named_parameters(): p[1].data = model_dict[p[0]]
for p in classifier.named_parameters(): p[1].data = class_dict[p[0]]
#wargs.start_epoch = eid + 1
else:
for p in nmtModel.parameters(): init_params(p, uniform=True)
for p in classifier.parameters(): init_params(p, uniform=True)
optim = Optim(
wargs.opt_mode, wargs.learning_rate, wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu
)
if wargs.gpu_id:
wlog('Push model onto GPU ... ')
nmtModel.cuda()
classifier.cuda()
else:
wlog('Push model onto CPU ... ')
nmtModel.cpu()
classifier.cuda()
nmtModel.classifier = classifier
wlog(nmtModel)
pcnt1 = len([p for p in nmtModel.parameters()])
pcnt2 = sum([p.nelement() for p in nmtModel.parameters()])
wlog('Parameters number: {}/{}'.format(pcnt1, pcnt2))
optim.init_optimizer(nmtModel.parameters())
#tor = Translator(nmtModel, src_vocab.idx2key, trg_vocab.idx2key)
#tor.trans_tests(tests_data, pre_dict['epoch'], pre_dict['batch'])
trainer = Trainer(nmtModel, src_vocab.idx2key, trg_vocab.idx2key, optim, trg_vocab_size)
dev_src0 = wargs.dir_data + 'dev.1k.zh0'
dev_trg0 = wargs.dir_data + 'dev.1k.en0'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src0, dev_trg0))
dev_src0, dev_trg0 = wrap_data(dev_src0, dev_trg0, src_vocab, trg_vocab)
wlog(len(train_src_tlst))
# add 1000 to train
train_all_chunks = (train_src_tlst, train_trg_tlst)
dh = DataHisto(train_all_chunks)
dev_src1 = wargs.dir_data + 'dev.1k.zh1'
dev_trg1 = wargs.dir_data + 'dev.1k.en1'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src1, dev_trg1))
dev_src1, dev_trg1 = wrap_data(dev_src1, dev_trg1, src_vocab, trg_vocab)
dev_src2 = wargs.dir_data + 'dev.1k.zh2'
dev_trg2 = wargs.dir_data + 'dev.1k.en2'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src2, dev_trg2))
dev_src2, dev_trg2 = wrap_data(dev_src2, dev_trg2, src_vocab, trg_vocab)
dev_src3 = wargs.dir_data + 'dev.1k.zh3'
dev_trg3 = wargs.dir_data + 'dev.1k.en3'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src3, dev_trg3))
dev_src3, dev_trg3 = wrap_data(dev_src3, dev_trg3, src_vocab, trg_vocab)
dev_src4 = wargs.dir_data + 'dev.1k.zh4'
dev_trg4 = wargs.dir_data + 'dev.1k.en4'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src4, dev_trg4))
dev_src4, dev_trg4 = wrap_data(dev_src4, dev_trg4, src_vocab, trg_vocab)
wlog(len(dev_src4+dev_src3+dev_src2+dev_src1+dev_src0))
dev_input = Input(dev_src4+dev_src3+dev_src2+dev_src1+dev_src0, dev_trg4+dev_trg3+dev_trg2+dev_trg1+dev_trg0, wargs.batch_size)
trainer.train(dh, dev_input, 0, batch_valid, tests_data, merge=True, name='DH_{}'.format('dev'))
'''
chunk_size = 1000
rand_ids = tc.randperm(len(train_src_tlst))[:chunk_size * 1000]
rand_ids = rand_ids.split(chunk_size)
#train_chunks = [(dev_src, dev_trg)]
train_chunks = []
for k in range(len(rand_ids)):
rand_id = rand_ids[k]
chunk_src_tlst = [train_src_tlst[i] for i in rand_id]
chunk_trg_tlst = [train_trg_tlst[i] for i in rand_id]
#wlog('Sentence-pairs count in training data: {}'.format(len(src_samples_train)))
#batch_train = Input(train_src_tlst, train_trg_tlst, wargs.batch_size)
#batch_train = Input(src_samples_train, trg_samples_train, wargs.batch_size)
train_chunks.append((chunk_src_tlst, chunk_trg_tlst))
chunk_D0 = train_chunks[0]
dh = DataHisto(chunk_D0)
c0_input = Input(chunk_D0[0], chunk_D0[1], wargs.batch_size)
trainer.train(dh, c0_input, 0, batch_valid, tests_data, merge=False, name='DH_{}'.format(0))
for k in range(1, len(train_chunks)):
wlog('*' * 30, False)
wlog(' Next Data {} '.format(k), False)
wlog('*' * 30)
chunk_Dk = train_chunks[k]
ck_input = Input(chunk_Dk[0], chunk_Dk[1], wargs.batch_size)
trainer.train(dh, ck_input, k, batch_valid, tests_data, merge=True, name='DH_{}'.format(k))
dh.add_batch_data(chunk_Dk)
'''
if tests_data and wargs.final_test:
bestModel = NMT()
classifier = Classifier(wargs.out_size, trg_vocab_size)
assert os.path.exists(wargs.best_model)
model_dict = tc.load(wargs.best_model)
best_model_dict = model_dict['model']
best_model_dict = {k: v for k, v in best_model_dict.items() if 'classifier' not in k}
bestModel.load_state_dict(best_model_dict)
classifier.load_state_dict(model_dict['class'])
if wargs.gpu_id:
wlog('Push NMT model onto GPU ... ')
bestModel.cuda()
classifier.cuda()
else:
wlog('Push NMT model onto CPU ... ')
bestModel.cpu()
classifier.cpu()
bestModel.classifier = classifier
tor = Translator(bestModel, src_vocab.idx2key, trg_vocab.idx2key)
tor.trans_tests(tests_data, model_dict['epoch'], model_dict['batch'])
def main():
init_dir(wargs.dir_model)
init_dir(wargs.dir_valid)
vocab_data = {}
train_srcD_file = wargs.src_vocab_from
wlog('\nPreparing source vocabulary from {} ... '.format(train_srcD_file))
src_vocab = extract_vocab(train_srcD_file, wargs.src_dict,
wargs.src_dict_size)
vocab_data['src'] = src_vocab
train_trgD_file = wargs.trg_vocab_from
wlog('\nPreparing target vocabulary from {} ... '.format(train_trgD_file))
trg_vocab = extract_vocab(train_trgD_file, wargs.trg_dict,
wargs.trg_dict_size)
vocab_data['trg'] = trg_vocab
train_src_file = wargs.train_src
train_trg_file = wargs.train_trg
wlog('\nPreparing training set from {} and {} ... '.format(
train_src_file, train_trg_file))
train_src_tlst, train_trg_tlst = wrap_data(train_src_file,
train_trg_file,
src_vocab,
trg_vocab,
max_seq_len=wargs.max_seq_len)
'''
list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...]
no padding
'''
'''
devs = {}
dev_src = wargs.val_tst_dir + wargs.val_prefix + '.src'
dev_trg = wargs.val_tst_dir + wargs.val_prefix + '.ref0'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src, dev_trg))
dev_src, dev_trg = wrap_data(dev_src, dev_trg, src_vocab, trg_vocab)
devs['src'], devs['trg'] = dev_src, dev_trg
'''
valid_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.val_prefix,
wargs.val_src_suffix)
wlog('\nPreparing validation set from {} ... '.format(valid_file))
valid_src_tlst, valid_src_lens = val_wrap_data(valid_file, src_vocab)
wlog('Sentence-pairs count in training data: {}'.format(
len(train_src_tlst)))
src_vocab_size, trg_vocab_size = vocab_data['src'].size(
), vocab_data['trg'].size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(
src_vocab_size, trg_vocab_size))
batch_train = Input(train_src_tlst, train_trg_tlst, wargs.batch_size)
batch_valid = Input(valid_src_tlst, None, 1, volatile=True)
tests_data = None
if wargs.tests_prefix is not None:
init_dir(wargs.dir_tests)
tests_data = {}
for prefix in wargs.tests_prefix:
init_dir(wargs.dir_tests + '/' + prefix)
test_file = '{}{}.{}'.format(wargs.val_tst_dir, prefix,
wargs.val_src_suffix)
wlog('Preparing test set from {} ... '.format(test_file))
test_src_tlst, _ = val_wrap_data(test_file, src_vocab)
tests_data[prefix] = Input(test_src_tlst, None, 1, volatile=True)
'''
# lookup_table on cpu to save memory
src_lookup_table = nn.Embedding(wargs.src_dict_size + 4,
wargs.src_wemb_size, padding_idx=utils.PAD).cpu()
trg_lookup_table = nn.Embedding(wargs.trg_dict_size + 4,
wargs.trg_wemb_size, padding_idx=utils.PAD).cpu()
wlog('Lookup table on CPU ... ')
wlog(src_lookup_table)
wlog(trg_lookup_table)
'''
sv = vocab_data['src'].idx2key
tv = vocab_data['trg'].idx2key
nmtModel = NMT(src_vocab_size, trg_vocab_size)
#classifier = Classifier(wargs.out_size, trg_vocab_size,
# nmtModel.decoder.trg_lookup_table if wargs.copy_trg_emb is True else None)
if wargs.pre_train:
assert os.path.exists(wargs.pre_train)
_dict = _load_model(wargs.pre_train)
# initializing parameters of interactive attention model
class_dict = None
if len(_dict) == 4: model_dict, eid, bid, optim = _dict
elif len(_dict) == 5:
model_dict, class_dict, eid, bid, optim = _dict
for name, param in nmtModel.named_parameters():
if name in model_dict:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(model_dict[name])
wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad,
name))
elif name.endswith('map_vocab.weight'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.weight'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
elif name.endswith('map_vocab.bias'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.bias'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
else:
init_params(param, name, True)
wargs.start_epoch = eid + 1
#tor = Translator(nmtModel, sv, tv)
#tor.trans_tests(tests_data, eid, bid)
else:
for n, p in nmtModel.named_parameters():
init_params(p, n, True)
#for n, p in classifier.named_parameters(): init_params(p, n, True)
optim = Optim(wargs.opt_mode,
wargs.learning_rate,
wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu)
if wargs.gpu_id:
nmtModel.cuda()
#classifier.cuda()
wlog('Push model onto GPU[{}] ... '.format(wargs.gpu_id[0]))
else:
nmtModel.cpu()
#classifier.cpu()
wlog('Push model onto CPU ... ')
#nmtModel.classifier = classifier
#nmtModel.decoder.map_vocab = classifier.map_vocab
'''
nmtModel.src_lookup_table = src_lookup_table
nmtModel.trg_lookup_table = trg_lookup_table
print nmtModel.src_lookup_table.weight.data.is_cuda
nmtModel.classifier.init_weights(nmtModel.trg_lookup_table)
'''
wlog(nmtModel)
wlog(optim)
pcnt1 = len([p for p in nmtModel.parameters()])
pcnt2 = sum([p.nelement() for p in nmtModel.parameters()])
wlog('Parameters number: {}/{}'.format(pcnt1, pcnt2))
optim.init_optimizer(nmtModel.parameters())
#tor = Translator(nmtModel, sv, tv, wargs.search_mode)
#tor.trans_tests(tests_data, pre_dict['epoch'], pre_dict['batch'])
trainer = Trainer(nmtModel, batch_train, vocab_data, optim, batch_valid,
tests_data)
trainer.train()
def main():
# Check if CUDA is available
if cuda.is_available():
wlog(
'CUDA is available, specify device by gpu_id argument (i.e. gpu_id=[3])'
)
else:
wlog('Warning: CUDA is not available, try CPU')
if wargs.gpu_id:
cuda.set_device(wargs.gpu_id[0])
wlog('Using GPU {}'.format(wargs.gpu_id[0]))
init_dir(wargs.dir_model)
init_dir(wargs.dir_valid)
'''
train_srcD_file = wargs.dir_data + 'train.10k.zh5'
wlog('\nPreparing source vocabulary from {} ... '.format(train_srcD_file))
src_vocab = extract_vocab(train_srcD_file, wargs.src_dict, wargs.src_dict_size)
train_trgD_file = wargs.dir_data + 'train.10k.en5'
wlog('\nPreparing target vocabulary from {} ... '.format(train_trgD_file))
trg_vocab = extract_vocab(train_trgD_file, wargs.trg_dict, wargs.trg_dict_size)
train_src_file = wargs.dir_data + 'train.10k.zh0'
train_trg_file = wargs.dir_data + 'train.10k.en0'
wlog('\nPreparing training set from {} and {} ... '.format(train_src_file, train_trg_file))
train_src_tlst, train_trg_tlst = wrap_data(train_src_file, train_trg_file, src_vocab, trg_vocab)
#list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...], no padding
wlog('Sentence-pairs count in training data: {}'.format(len(train_src_tlst)))
src_vocab_size, trg_vocab_size = src_vocab.size(), trg_vocab.size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(src_vocab_size, trg_vocab_size))
batch_train = Input(train_src_tlst, train_trg_tlst, wargs.batch_size)
'''
src = os.path.join(
wargs.dir_data, '{}.{}'.format(wargs.train_prefix,
wargs.train_src_suffix))
trg = os.path.join(
wargs.dir_data, '{}.{}'.format(wargs.train_prefix,
wargs.train_trg_suffix))
vocabs = {}
wlog('\nPreparing source vocabulary from {} ... '.format(src))
src_vocab = extract_vocab(src, wargs.src_dict, wargs.src_dict_size)
wlog('\nPreparing target vocabulary from {} ... '.format(trg))
trg_vocab = extract_vocab(trg, wargs.trg_dict, wargs.trg_dict_size)
src_vocab_size, trg_vocab_size = src_vocab.size(), trg_vocab.size()
wlog('Vocabulary size: |source|={}, |target|={}'.format(
src_vocab_size, trg_vocab_size))
vocabs['src'], vocabs['trg'] = src_vocab, trg_vocab
wlog('\nPreparing training set from {} and {} ... '.format(src, trg))
trains = {}
train_src_tlst, train_trg_tlst = wrap_data(wargs.dir_data,
wargs.train_prefix,
wargs.train_src_suffix,
wargs.train_trg_suffix,
src_vocab,
trg_vocab,
max_seq_len=wargs.max_seq_len)
'''
list [torch.LongTensor (sentence), torch.LongTensor, torch.LongTensor, ...]
no padding
'''
batch_train = Input(train_src_tlst,
train_trg_tlst,
wargs.batch_size,
batch_sort=True)
wlog('Sentence-pairs count in training data: {}'.format(
len(train_src_tlst)))
batch_valid = None
if wargs.val_prefix is not None:
val_src_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.val_prefix,
wargs.val_src_suffix)
val_trg_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.val_prefix,
wargs.val_ref_suffix)
wlog('\nPreparing validation set from {} and {} ... '.format(
val_src_file, val_trg_file))
valid_src_tlst, valid_trg_tlst = wrap_data(
wargs.val_tst_dir,
wargs.val_prefix,
wargs.val_src_suffix,
wargs.val_ref_suffix,
src_vocab,
trg_vocab,
shuffle=False,
sort_data=False,
max_seq_len=wargs.dev_max_seq_len)
batch_valid = Input(valid_src_tlst,
valid_trg_tlst,
1,
volatile=True,
batch_sort=False)
batch_tests = None
if wargs.tests_prefix is not None:
assert isinstance(wargs.tests_prefix,
list), 'Test files should be list.'
init_dir(wargs.dir_tests)
batch_tests = {}
for prefix in wargs.tests_prefix:
init_dir(wargs.dir_tests + '/' + prefix)
test_file = '{}{}.{}'.format(wargs.val_tst_dir, prefix,
wargs.val_src_suffix)
wlog('\nPreparing test set from {} ... '.format(test_file))
test_src_tlst, _ = wrap_tst_data(test_file, src_vocab)
batch_tests[prefix] = Input(test_src_tlst,
None,
1,
volatile=True,
batch_sort=False)
wlog('\n## Finish to Prepare Dataset ! ##\n')
nmtModel = NMT(src_vocab_size, trg_vocab_size)
if wargs.pre_train is not None:
assert os.path.exists(wargs.pre_train)
_dict = _load_model(wargs.pre_train)
# initializing parameters of interactive attention model
class_dict = None
if len(_dict) == 4: model_dict, eid, bid, optim = _dict
elif len(_dict) == 5:
model_dict, class_dict, eid, bid, optim = _dict
for name, param in nmtModel.named_parameters():
if name in model_dict:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(model_dict[name])
wlog('{:7} -> grad {}\t{}'.format('Model', param.requires_grad,
name))
elif name.endswith('map_vocab.weight'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.weight'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
elif name.endswith('map_vocab.bias'):
if class_dict is not None:
param.requires_grad = not wargs.fix_pre_params
param.data.copy_(class_dict['map_vocab.bias'])
wlog('{:7} -> grad {}\t{}'.format('Model',
param.requires_grad,
name))
else:
init_params(param, name, True)
wargs.start_epoch = eid + 1
else:
for n, p in nmtModel.named_parameters():
init_params(p, n, True)
optim = Optim(wargs.opt_mode,
wargs.learning_rate,
wargs.max_grad_norm,
learning_rate_decay=wargs.learning_rate_decay,
start_decay_from=wargs.start_decay_from,
last_valid_bleu=wargs.last_valid_bleu)
optim.init_optimizer(nmtModel.parameters())
if wargs.gpu_id:
wlog('Push model onto GPU {} ... '.format(wargs.gpu_id), 0)
nmtModel.cuda()
else:
wlog('Push model onto CPU ... ', 0)
nmtModel.cpu()
wlog('done.')
wlog(nmtModel)
wlog(optim)
pcnt1 = len([p for p in nmtModel.parameters()])
pcnt2 = sum([p.nelement() for p in nmtModel.parameters()])
wlog('Parameters number: {}/{}'.format(pcnt1, pcnt2))
trainer = Trainer(nmtModel,
src_vocab.idx2key,
trg_vocab.idx2key,
optim,
trg_vocab_size,
valid_data=batch_valid,
tests_data=batch_tests)
# add 1000 to train
train_all_chunks = (train_src_tlst, train_trg_tlst)
dh = DataHisto(train_all_chunks)
'''
dev_src0 = wargs.dir_data + 'dev.1k.zh0'
dev_trg0 = wargs.dir_data + 'dev.1k.en0'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src0, dev_trg0))
dev_src0, dev_trg0 = wrap_data(dev_src0, dev_trg0, src_vocab, trg_vocab)
wlog(len(train_src_tlst))
dev_src1 = wargs.dir_data + 'dev.1k.zh1'
dev_trg1 = wargs.dir_data + 'dev.1k.en1'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src1, dev_trg1))
dev_src1, dev_trg1 = wrap_data(dev_src1, dev_trg1, src_vocab, trg_vocab)
dev_src2 = wargs.dir_data + 'dev.1k.zh2'
dev_trg2 = wargs.dir_data + 'dev.1k.en2'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src2, dev_trg2))
dev_src2, dev_trg2 = wrap_data(dev_src2, dev_trg2, src_vocab, trg_vocab)
dev_src3 = wargs.dir_data + 'dev.1k.zh3'
dev_trg3 = wargs.dir_data + 'dev.1k.en3'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src3, dev_trg3))
dev_src3, dev_trg3 = wrap_data(dev_src3, dev_trg3, src_vocab, trg_vocab)
dev_src4 = wargs.dir_data + 'dev.1k.zh4'
dev_trg4 = wargs.dir_data + 'dev.1k.en4'
wlog('\nPreparing dev set for tuning from {} and {} ... '.format(dev_src4, dev_trg4))
dev_src4, dev_trg4 = wrap_data(dev_src4, dev_trg4, src_vocab, trg_vocab)
wlog(len(dev_src4+dev_src3+dev_src2+dev_src1+dev_src0))
batch_dev = Input(dev_src4+dev_src3+dev_src2+dev_src1+dev_src0, dev_trg4+dev_trg3+dev_trg2+dev_trg1+dev_trg0, wargs.batch_size)
'''
batch_dev = None
assert wargs.dev_prefix is not None, 'Requires development to tuning.'
dev_src_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.dev_prefix,
wargs.val_src_suffix)
dev_trg_file = '{}{}.{}'.format(wargs.val_tst_dir, wargs.dev_prefix,
wargs.val_ref_suffix)
wlog('\nPreparing dev set from {} and {} ... '.format(
dev_src_file, dev_trg_file))
valid_src_tlst, valid_trg_tlst = wrap_data(
wargs.val_tst_dir,
wargs.dev_prefix,
wargs.val_src_suffix,
wargs.val_ref_suffix,
src_vocab,
trg_vocab,
shuffle=True,
sort_data=True,
max_seq_len=wargs.dev_max_seq_len)
batch_dev = Input(valid_src_tlst,
valid_trg_tlst,
wargs.batch_size,
batch_sort=True)
trainer.train(dh, batch_dev, 0, merge=True, name='DH_{}'.format('dev'))
'''