File size: 4,021 Bytes
5000658 |
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 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 |
# SPDX-FileCopyrightText: Copyright (c) 2022-2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
#
# 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 ..functional import group_norm, layer_norm, rms_norm
from ..module import Module
from ..parameter import Parameter
class LayerNorm(Module):
def __init__(self,
normalized_shape,
eps=1e-05,
elementwise_affine=True,
bias=True,
dtype=None):
super().__init__()
if isinstance(normalized_shape, int):
normalized_shape = (normalized_shape, )
self.normalized_shape = tuple(normalized_shape)
self.elementwise_affine = elementwise_affine
if self.elementwise_affine:
self.weight = Parameter(shape=self.normalized_shape, dtype=dtype)
if bias:
self.bias = Parameter(shape=self.normalized_shape, dtype=dtype)
else:
self.register_parameter('bias', None)
else:
self.register_parameter('weight', None)
self.register_parameter('bias', None)
self.eps = eps
self.dtype = dtype
def forward(self, x):
weight = 1. if self.weight is None else self.weight.value
bias = 0. if self.bias is None else self.bias.value
return layer_norm(x, self.normalized_shape, weight, bias, self.eps)
class RmsNorm(Module):
def __init__(self,
normalized_shape,
num_groups=1,
eps=1e-06,
elementwise_affine=True,
dtype=None):
super().__init__()
if isinstance(normalized_shape, int):
normalized_shape = (normalized_shape, )
self.normalized_shape = tuple(normalized_shape)
self.elementwise_affine = elementwise_affine
self.num_groups = num_groups
num_channels = normalized_shape[-1]
if num_channels % num_groups != 0:
raise ValueError('num_channels must be divisible by num_groups')
if self.elementwise_affine:
self.weight = Parameter(shape=self.normalized_shape, dtype=dtype)
else:
self.register_parameter('weight', None)
self.eps = eps
self.dtype = dtype
def forward(self, x):
weight = None if self.weight is None else self.weight.value
return rms_norm(x, self.normalized_shape, self.num_groups, weight,
self.eps)
class GroupNorm(Module):
def __init__(self,
num_groups,
num_channels,
eps=1e-05,
affine=True,
dtype=None):
super().__init__()
if num_channels % num_groups != 0:
raise ValueError('num_channels must be divisible by num_groups')
self.num_groups = num_groups
self.num_channels = num_channels
self.affine = affine
if self.affine:
self.weight = Parameter(shape=(self.num_channels, ), dtype=dtype)
self.bias = Parameter(shape=(self.num_channels, ), dtype=dtype)
else:
self.register_parameter('weight', None)
self.register_parameter('bias', None)
self.eps = eps
def forward(self, x):
weight = None if self.weight is None else self.weight.value
bias = None if self.bias is None else self.bias.value
return group_norm(x, self.num_groups, weight, bias, self.eps)
|