File size: 12,055 Bytes
908a1ab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
# Modified from https://github.com/mseitzer/pytorch-fid/blob/master/pytorch_fid/inception.py # noqa: E501
# For FID metric

import os
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.utils.model_zoo import load_url
from torchvision import models

# Inception weights ported to Pytorch from
# http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz
FID_WEIGHTS_URL = 'https://github.com/mseitzer/pytorch-fid/releases/download/fid_weights/pt_inception-2015-12-05-6726825d.pth'  # noqa: E501
LOCAL_FID_WEIGHTS = 'experiments/pretrained_models/pt_inception-2015-12-05-6726825d.pth'  # noqa: E501


class InceptionV3(nn.Module):
    """Pretrained InceptionV3 network returning feature maps"""

    # Index of default block of inception to return,
    # corresponds to output of final average pooling
    DEFAULT_BLOCK_INDEX = 3

    # Maps feature dimensionality to their output blocks indices
    BLOCK_INDEX_BY_DIM = {
        64: 0,  # First max pooling features
        192: 1,  # Second max pooling features
        768: 2,  # Pre-aux classifier features
        2048: 3  # Final average pooling features
    }

    def __init__(self,
                 output_blocks=(DEFAULT_BLOCK_INDEX),
                 resize_input=True,
                 normalize_input=True,
                 requires_grad=False,
                 use_fid_inception=True):
        """Build pretrained InceptionV3.

        Args:
            output_blocks (list[int]): Indices of blocks to return features of.
                Possible values are:
                - 0: corresponds to output of first max pooling
                - 1: corresponds to output of second max pooling
                - 2: corresponds to output which is fed to aux classifier
                - 3: corresponds to output of final average pooling
            resize_input (bool): If true, bilinearly resizes input to width and
                height 299 before feeding input to model. As the network
                without fully connected layers is fully convolutional, it
                should be able to handle inputs of arbitrary size, so resizing
                might not be strictly needed. Default: True.
            normalize_input (bool): If true, scales the input from range (0, 1)
                to the range the pretrained Inception network expects,
                namely (-1, 1). Default: True.
            requires_grad (bool): If true, parameters of the model require
                gradients. Possibly useful for finetuning the network.
                Default: False.
            use_fid_inception (bool): If true, uses the pretrained Inception
                model used in Tensorflow's FID implementation.
                If false, uses the pretrained Inception model available in
                torchvision. The FID Inception model has different weights
                and a slightly different structure from torchvision's
                Inception model. If you want to compute FID scores, you are
                strongly advised to set this parameter to true to get
                comparable results. Default: True.
        """
        super(InceptionV3, self).__init__()

        self.resize_input = resize_input
        self.normalize_input = normalize_input
        self.output_blocks = sorted(output_blocks)
        self.last_needed_block = max(output_blocks)

        assert self.last_needed_block <= 3, ('Last possible output block index is 3')

        self.blocks = nn.ModuleList()

        if use_fid_inception:
            inception = fid_inception_v3()
        else:
            try:
                inception = models.inception_v3(pretrained=True, init_weights=False)
            except TypeError:
                # pytorch < 1.5 does not have init_weights for inception_v3
                inception = models.inception_v3(pretrained=True)

        # Block 0: input to maxpool1
        block0 = [
            inception.Conv2d_1a_3x3, inception.Conv2d_2a_3x3, inception.Conv2d_2b_3x3,
            nn.MaxPool2d(kernel_size=3, stride=2)
        ]
        self.blocks.append(nn.Sequential(*block0))

        # Block 1: maxpool1 to maxpool2
        if self.last_needed_block >= 1:
            block1 = [inception.Conv2d_3b_1x1, inception.Conv2d_4a_3x3, nn.MaxPool2d(kernel_size=3, stride=2)]
            self.blocks.append(nn.Sequential(*block1))

        # Block 2: maxpool2 to aux classifier
        if self.last_needed_block >= 2:
            block2 = [
                inception.Mixed_5b,
                inception.Mixed_5c,
                inception.Mixed_5d,
                inception.Mixed_6a,
                inception.Mixed_6b,
                inception.Mixed_6c,
                inception.Mixed_6d,
                inception.Mixed_6e,
            ]
            self.blocks.append(nn.Sequential(*block2))

        # Block 3: aux classifier to final avgpool
        if self.last_needed_block >= 3:
            block3 = [
                inception.Mixed_7a, inception.Mixed_7b, inception.Mixed_7c,
                nn.AdaptiveAvgPool2d(output_size=(1, 1))
            ]
            self.blocks.append(nn.Sequential(*block3))

        for param in self.parameters():
            param.requires_grad = requires_grad

    def forward(self, x):
        """Get Inception feature maps.

        Args:
            x (Tensor): Input tensor of shape (b, 3, h, w).
                Values are expected to be in range (-1, 1). You can also input
                (0, 1) with setting normalize_input = True.

        Returns:
            list[Tensor]: Corresponding to the selected output block, sorted
            ascending by index.
        """
        output = []

        if self.resize_input:
            x = F.interpolate(x, size=(299, 299), mode='bilinear', align_corners=False)

        if self.normalize_input:
            x = 2 * x - 1  # Scale from range (0, 1) to range (-1, 1)

        for idx, block in enumerate(self.blocks):
            x = block(x)
            if idx in self.output_blocks:
                output.append(x)

            if idx == self.last_needed_block:
                break

        return output


def fid_inception_v3():
    """Build pretrained Inception model for FID computation.

    The Inception model for FID computation uses a different set of weights
    and has a slightly different structure than torchvision's Inception.

    This method first constructs torchvision's Inception and then patches the
    necessary parts that are different in the FID Inception model.
    """
    try:
        inception = models.inception_v3(num_classes=1008, aux_logits=False, pretrained=False, init_weights=False)
    except TypeError:
        # pytorch < 1.5 does not have init_weights for inception_v3
        inception = models.inception_v3(num_classes=1008, aux_logits=False, pretrained=False)

    inception.Mixed_5b = FIDInceptionA(192, pool_features=32)
    inception.Mixed_5c = FIDInceptionA(256, pool_features=64)
    inception.Mixed_5d = FIDInceptionA(288, pool_features=64)
    inception.Mixed_6b = FIDInceptionC(768, channels_7x7=128)
    inception.Mixed_6c = FIDInceptionC(768, channels_7x7=160)
    inception.Mixed_6d = FIDInceptionC(768, channels_7x7=160)
    inception.Mixed_6e = FIDInceptionC(768, channels_7x7=192)
    inception.Mixed_7b = FIDInceptionE_1(1280)
    inception.Mixed_7c = FIDInceptionE_2(2048)

    if os.path.exists(LOCAL_FID_WEIGHTS):
        state_dict = torch.load(LOCAL_FID_WEIGHTS, map_location=lambda storage, loc: storage)
    else:
        state_dict = load_url(FID_WEIGHTS_URL, progress=True)

    inception.load_state_dict(state_dict)
    return inception


class FIDInceptionA(models.inception.InceptionA):
    """InceptionA block patched for FID computation"""

    def __init__(self, in_channels, pool_features):
        super(FIDInceptionA, self).__init__(in_channels, pool_features)

    def forward(self, x):
        branch1x1 = self.branch1x1(x)

        branch5x5 = self.branch5x5_1(x)
        branch5x5 = self.branch5x5_2(branch5x5)

        branch3x3dbl = self.branch3x3dbl_1(x)
        branch3x3dbl = self.branch3x3dbl_2(branch3x3dbl)
        branch3x3dbl = self.branch3x3dbl_3(branch3x3dbl)

        # Patch: Tensorflow's average pool does not use the padded zero's in
        # its average calculation
        branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1, count_include_pad=False)
        branch_pool = self.branch_pool(branch_pool)

        outputs = [branch1x1, branch5x5, branch3x3dbl, branch_pool]
        return torch.cat(outputs, 1)


class FIDInceptionC(models.inception.InceptionC):
    """InceptionC block patched for FID computation"""

    def __init__(self, in_channels, channels_7x7):
        super(FIDInceptionC, self).__init__(in_channels, channels_7x7)

    def forward(self, x):
        branch1x1 = self.branch1x1(x)

        branch7x7 = self.branch7x7_1(x)
        branch7x7 = self.branch7x7_2(branch7x7)
        branch7x7 = self.branch7x7_3(branch7x7)

        branch7x7dbl = self.branch7x7dbl_1(x)
        branch7x7dbl = self.branch7x7dbl_2(branch7x7dbl)
        branch7x7dbl = self.branch7x7dbl_3(branch7x7dbl)
        branch7x7dbl = self.branch7x7dbl_4(branch7x7dbl)
        branch7x7dbl = self.branch7x7dbl_5(branch7x7dbl)

        # Patch: Tensorflow's average pool does not use the padded zero's in
        # its average calculation
        branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1, count_include_pad=False)
        branch_pool = self.branch_pool(branch_pool)

        outputs = [branch1x1, branch7x7, branch7x7dbl, branch_pool]
        return torch.cat(outputs, 1)


class FIDInceptionE_1(models.inception.InceptionE):
    """First InceptionE block patched for FID computation"""

    def __init__(self, in_channels):
        super(FIDInceptionE_1, self).__init__(in_channels)

    def forward(self, x):
        branch1x1 = self.branch1x1(x)

        branch3x3 = self.branch3x3_1(x)
        branch3x3 = [
            self.branch3x3_2a(branch3x3),
            self.branch3x3_2b(branch3x3),
        ]
        branch3x3 = torch.cat(branch3x3, 1)

        branch3x3dbl = self.branch3x3dbl_1(x)
        branch3x3dbl = self.branch3x3dbl_2(branch3x3dbl)
        branch3x3dbl = [
            self.branch3x3dbl_3a(branch3x3dbl),
            self.branch3x3dbl_3b(branch3x3dbl),
        ]
        branch3x3dbl = torch.cat(branch3x3dbl, 1)

        # Patch: Tensorflow's average pool does not use the padded zero's in
        # its average calculation
        branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1, count_include_pad=False)
        branch_pool = self.branch_pool(branch_pool)

        outputs = [branch1x1, branch3x3, branch3x3dbl, branch_pool]
        return torch.cat(outputs, 1)


class FIDInceptionE_2(models.inception.InceptionE):
    """Second InceptionE block patched for FID computation"""

    def __init__(self, in_channels):
        super(FIDInceptionE_2, self).__init__(in_channels)

    def forward(self, x):
        branch1x1 = self.branch1x1(x)

        branch3x3 = self.branch3x3_1(x)
        branch3x3 = [
            self.branch3x3_2a(branch3x3),
            self.branch3x3_2b(branch3x3),
        ]
        branch3x3 = torch.cat(branch3x3, 1)

        branch3x3dbl = self.branch3x3dbl_1(x)
        branch3x3dbl = self.branch3x3dbl_2(branch3x3dbl)
        branch3x3dbl = [
            self.branch3x3dbl_3a(branch3x3dbl),
            self.branch3x3dbl_3b(branch3x3dbl),
        ]
        branch3x3dbl = torch.cat(branch3x3dbl, 1)

        # Patch: The FID Inception model uses max pooling instead of average
        # pooling. This is likely an error in this specific Inception
        # implementation, as other Inception models use average pooling here
        # (which matches the description in the paper).
        branch_pool = F.max_pool2d(x, kernel_size=3, stride=1, padding=1)
        branch_pool = self.branch_pool(branch_pool)

        outputs = [branch1x1, branch3x3, branch3x3dbl, branch_pool]
        return torch.cat(outputs, 1)