# MIT License
#
# Copyright (c) 2021 Soohwan Kim and Sangchun Ha and Soyoung Cho
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import torch.nn as nn
from torch import Tensor
from openspeech.modules.wrapper import Linear
[docs]class PositionwiseFeedForward(nn.Module):
"""
Position-wise Feedforward Networks proposed in "Attention Is All You Need".
Fully connected feed-forward network, which is applied to each position separately and identically.
This consists of two linear transformations with a ReLU activation in between.
Another way of describing this is as two convolutions with kernel size 1.
"""
def __init__(self, d_model: int = 512, d_ff: int = 2048, dropout_p: float = 0.3) -> None:
super(PositionwiseFeedForward, self).__init__()
self.feed_forward = nn.Sequential(
Linear(d_model, d_ff),
nn.Dropout(dropout_p),
nn.ReLU(),
Linear(d_ff, d_model),
nn.Dropout(dropout_p),
)
def forward(self, inputs: Tensor) -> Tensor:
return self.feed_forward(inputs)