1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
|
- # Copyright (c) 2017-present, Facebook, Inc.
- # All rights reserved.
- #
- # This source code is licensed under the license found in the LICENSE file in
- # the root directory of this source tree. An additional grant of patent rights
- # can be found in the PATENTS file in the same directory.
- import math
- from fairseq import utils
- from . import FairseqCriterion, register_criterion
- @register_criterion('label_smoothed_cross_entropy')
- class LabelSmoothedCrossEntropyCriterion(FairseqCriterion):
- def __init__(self, args, src_dict, dst_dict):
- super().__init__(args, src_dict, dst_dict)
- self.eps = args.label_smoothing
- @staticmethod
- def add_args(parser):
- """Add criterion-specific arguments to the parser."""
- parser.add_argument('--label-smoothing', default=0., type=float, metavar='D',
- help='epsilon for label smoothing, 0 means no label smoothing')
- def forward(self, model, sample, reduce=True):
- """Compute the loss for the given sample.
- Returns a tuple with three elements:
- 1) the loss, as a Variable
- 2) the sample size, which is used as the denominator for the gradient
- 3) logging outputs to display while training
- """
- net_output = model(**sample['net_input'])
- lprobs = model.get_normalized_probs(net_output, log_probs=True)
- lprobs = lprobs.view(-1, lprobs.size(-1))
- target = model.get_targets(sample, net_output).view(-1, 1)
- non_pad_mask = target.ne(self.padding_idx)
- nll_loss = -lprobs.gather(dim=-1, index=target)[non_pad_mask]
- smooth_loss = -lprobs.sum(dim=-1, keepdim=True)[non_pad_mask]
- if reduce:
- nll_loss = nll_loss.sum()
- smooth_loss = smooth_loss.sum()
- eps_i = self.eps / lprobs.size(-1)
- loss = (1. - self.eps) * nll_loss + eps_i * smooth_loss
- sample_size = sample['target'].size(0) if self.args.sentence_avg else sample['ntokens']
- logging_output = {
- 'loss': utils.item(loss.data) if reduce else loss.data,
- 'nll_loss': utils.item(nll_loss.data) if reduce else loss.data,
- 'ntokens': sample['ntokens'],
- 'sample_size': sample_size,
- }
- return loss, sample_size, logging_output
- @staticmethod
- def aggregate_logging_outputs(logging_outputs):
- """Aggregate logging outputs from data parallel training."""
- ntokens = sum(log.get('ntokens', 0) for log in logging_outputs)
- sample_size = sum(log.get('sample_size', 0) for log in logging_outputs)
- return {
- 'loss': sum(log.get('loss', 0) for log in logging_outputs) / sample_size / math.log(2),
- 'nll_loss': sum(log.get('nll_loss', 0) for log in logging_outputs) / ntokens / math.log(2),
- 'sample_size': sample_size,
- }
|