# Copyright 2022 Cerebras Systems.
#
# 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.
# coding=utf-8
#
# This code is adapted from
# https://github.com/huggingface/transformers/blob/master/src/transformers/models/bert/modeling_bert.py
#
# Copyright 2022 Cerebras Systems.
#
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
#
# 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 torch
import torch.nn as nn
from cerebras.modelzoo.common.registry import registry
from cerebras.modelzoo.common.utils.model.transformer_utils import smooth_loss
[docs]@registry.register_loss("BertPretrainModelLoss")
class BertPretrainModelLoss(nn.Module):
[docs] def __init__(
self,
disable_nsp=False,
mlm_loss_weight=1.0,
label_smoothing=0.0,
):
super(BertPretrainModelLoss, self).__init__()
self.disable_nsp = disable_nsp
self.mlm_loss_weight = mlm_loss_weight
self.label_smoothing = label_smoothing
def forward(
self,
mlm_logits,
vocab_size,
mlm_labels,
nsp_logits,
nsp_labels,
mlm_weights,
mlm_loss_scale=None,
):
mlm_loss_fn = nn.CrossEntropyLoss(reduction="none")
if mlm_loss_scale is not None:
mlm_weights = mlm_weights * mlm_loss_scale
mlm_loss = mlm_loss_fn(
mlm_logits.view(-1, vocab_size),
mlm_labels.view(-1).long(),
)
if self.label_smoothing > 0.0 and self.training:
# Calculate loss correction for label smoothing
mlm_loss = smooth_loss(
mlm_logits, mlm_loss, self.label_smoothing, vocab_size
)
mlm_loss *= mlm_weights.view(-1)
mlm_loss = torch.sum(mlm_loss) / mlm_labels.shape[0]
if mlm_loss_scale is None:
mlm_loss *= self.mlm_loss_weight
total_loss = mlm_loss
if not self.disable_nsp:
nsp_loss_fn = nn.CrossEntropyLoss()
nsp_loss = nsp_loss_fn(
nsp_logits.view(-1, 2),
nsp_labels.view(-1).long(),
)
total_loss += nsp_loss
return total_loss