forked from PaddlePaddle/PaddleOCR
-
Notifications
You must be signed in to change notification settings - Fork 0
/
rec_nrtr_loss.py
30 lines (27 loc) · 1.06 KB
/
rec_nrtr_loss.py
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
import paddle
from paddle import nn
import paddle.nn.functional as F
class NRTRLoss(nn.Layer):
def __init__(self, smoothing=True, **kwargs):
super(NRTRLoss, self).__init__()
self.loss_func = nn.CrossEntropyLoss(reduction='mean', ignore_index=0)
self.smoothing = smoothing
def forward(self, pred, batch):
pred = pred.reshape([-1, pred.shape[2]])
max_len = batch[2].max()
tgt = batch[1][:, 1:2 + max_len]
tgt = tgt.reshape([-1])
if self.smoothing:
eps = 0.1
n_class = pred.shape[1]
one_hot = F.one_hot(tgt, pred.shape[1])
one_hot = one_hot * (1 - eps) + (1 - one_hot) * eps / (n_class - 1)
log_prb = F.log_softmax(pred, axis=1)
non_pad_mask = paddle.not_equal(
tgt, paddle.zeros(
tgt.shape, dtype=tgt.dtype))
loss = -(one_hot * log_prb).sum(axis=1)
loss = loss.masked_select(non_pad_mask).mean()
else:
loss = self.loss_func(pred, tgt)
return {'loss': loss}