You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

77 lines
2.8 KiB
Python

# copyright (c) 2023 PaddlePaddle Authors. All Rights Reserve.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import paddle
from paddle import nn
import paddle.nn.functional as F
class CPPDLoss(nn.Layer):
def __init__(
self, smoothing=False, ignore_index=100, sideloss_weight=1.0, **kwargs
):
super(CPPDLoss, self).__init__()
self.edge_ce = nn.CrossEntropyLoss(reduction="mean", ignore_index=ignore_index)
self.char_node_ce = nn.CrossEntropyLoss(reduction="mean")
self.pos_node_ce = nn.BCEWithLogitsLoss(reduction="mean")
self.smoothing = smoothing
self.ignore_index = ignore_index
self.sideloss_weight = sideloss_weight
def label_smoothing_ce(self, preds, targets):
non_pad_mask = paddle.not_equal(
targets,
paddle.zeros(targets.shape, dtype=targets.dtype) + self.ignore_index,
)
tgts = paddle.where(
targets
== (paddle.zeros(targets.shape, dtype=targets.dtype) + self.ignore_index),
paddle.zeros(targets.shape, dtype=targets.dtype),
targets,
)
eps = 0.1
n_class = preds.shape[1]
one_hot = F.one_hot(tgts, preds.shape[1])
one_hot = one_hot * (1 - eps) + (1 - one_hot) * eps / (n_class - 1)
log_prb = F.log_softmax(preds, axis=1)
loss = -(one_hot * log_prb).sum(axis=1)
loss = loss.masked_select(non_pad_mask).mean()
return loss
def forward(self, pred, batch):
node_feats, edge_feats = pred
node_tgt = batch[2]
char_tgt = batch[1]
loss_char_node = self.char_node_ce(
node_feats[0].flatten(0, 1), node_tgt[:, :-26].flatten(0, 1)
)
loss_pos_node = self.pos_node_ce(
node_feats[1].flatten(0, 1), node_tgt[:, -26:].flatten(0, 1).cast("float32")
)
loss_node = loss_char_node + loss_pos_node
edge_feats = edge_feats.flatten(0, 1)
char_tgt = char_tgt.flatten(0, 1)
if self.smoothing:
loss_edge = self.label_smoothing_ce(edge_feats, char_tgt)
else:
loss_edge = self.edge_ce(edge_feats, char_tgt)
return {
"loss": self.sideloss_weight * loss_node + loss_edge,
"loss_node": self.sideloss_weight * loss_node,
"loss_edge": loss_edge,
}