File size: 3,604 Bytes
a89d9fd |
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 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 |
# copyright (c) 2021 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.
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import paddle
from paddle import nn
from paddle.nn import functional as F
class TableAttentionLoss(nn.Layer):
def __init__(self, structure_weight, loc_weight, **kwargs):
super(TableAttentionLoss, self).__init__()
self.loss_func = nn.CrossEntropyLoss(weight=None, reduction='none')
self.structure_weight = structure_weight
self.loc_weight = loc_weight
def forward(self, predicts, batch):
structure_probs = predicts['structure_probs']
structure_targets = batch[1].astype("int64")
structure_targets = structure_targets[:, 1:]
structure_probs = paddle.reshape(structure_probs,
[-1, structure_probs.shape[-1]])
structure_targets = paddle.reshape(structure_targets, [-1])
structure_loss = self.loss_func(structure_probs, structure_targets)
structure_loss = paddle.mean(structure_loss) * self.structure_weight
loc_preds = predicts['loc_preds']
loc_targets = batch[2].astype("float32")
loc_targets_mask = batch[3].astype("float32")
loc_targets = loc_targets[:, 1:, :]
loc_targets_mask = loc_targets_mask[:, 1:, :]
loc_loss = F.mse_loss(loc_preds * loc_targets_mask,
loc_targets) * self.loc_weight
total_loss = structure_loss + loc_loss
return {
'loss': total_loss,
"structure_loss": structure_loss,
"loc_loss": loc_loss
}
class SLALoss(nn.Layer):
def __init__(self, structure_weight, loc_weight, loc_loss='mse', **kwargs):
super(SLALoss, self).__init__()
self.loss_func = nn.CrossEntropyLoss(weight=None, reduction='mean')
self.structure_weight = structure_weight
self.loc_weight = loc_weight
self.loc_loss = loc_loss
self.eps = 1e-12
def forward(self, predicts, batch):
structure_probs = predicts['structure_probs']
structure_targets = batch[1].astype("int64")
structure_targets = structure_targets[:, 1:]
structure_loss = self.loss_func(structure_probs, structure_targets)
structure_loss = paddle.mean(structure_loss) * self.structure_weight
loc_preds = predicts['loc_preds']
loc_targets = batch[2].astype("float32")
loc_targets_mask = batch[3].astype("float32")
loc_targets = loc_targets[:, 1:, :]
loc_targets_mask = loc_targets_mask[:, 1:, :]
loc_loss = F.smooth_l1_loss(
loc_preds * loc_targets_mask,
loc_targets * loc_targets_mask,
reduction='sum') * self.loc_weight
loc_loss = loc_loss / (loc_targets_mask.sum() + self.eps)
total_loss = structure_loss + loc_loss
return {
'loss': total_loss,
"structure_loss": structure_loss,
"loc_loss": loc_loss
}
|