“”"Contains the Inception V3 model, which is used for inference ONLY.
This file is mostly borrowed from torchvision/models/inception.py.
Inception model is widely used to compute FID or IS metric for evaluating
generative models. However, the pre-trained models from torchvision is slightly
different from the TensorFlow version
http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz
which is used by the official FID implementation
https://github.com/bioinf-jku/TTUR
In particular:
(1) The number of classes in TensorFlow model is 1008 instead of 1000.
(2) The avg_pool() layers in TensorFlow model does not include the padded zero.
(3) The last Inception E Block in TensorFlow model use max_pool() instead of
avg_pool().
Hence, to align the evaluation results with those from TensorFlow
implementation, we modified the inception model to support both versions. Please
use align_tf argument to control the version.
“”"
import warnings
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.distributed as dist
from utils.misc import download_url
all = [‘InceptionModel’]
_MODEL_URL_SHA256 = {
# This model is provided by torchvision, which is ported from TensorFlow.
‘torchvision_official’: (
‘https://download.pytorch.org/models/inception_v3_google-1a9a5a14.pth’,
‘1a9a5a14f40645a370184bd54f4e8e631351e71399112b43ad0294a79da290c8’ # hash sha256
),
# This model is provided by https://github.com/mseitzer/pytorch-fid
'tf_inception_v3': ('https://github.com/mseitzer/pytorch-fid/releases/download/fid_weights/pt_inception-2015-12-05-6726825d.pth','6726825d0af5f729cebd5821db510b11b1cfad8faad88a03f1befd49fb9129b2' # hash sha256
)
}
class InceptionModel(object):
“”"Defines the Inception (V3) model.
This is a static class, which is used to avoid this model to be built
repeatedly. Consequently, this model is particularly used for inference,
like computing FID. If training is required, please use the model from
`torchvision.models` or implement by yourself.NOTE: The pre-trained model assumes the inputs to be with `RGB` channel
order and pixel range [-1, 1], and will also resize the images to shape
[299, 299] automatically. If your input is normalized by subtracting
(0.485, 0.456, 0.406) and dividing (0.229, 0.224, 0.225), please use
`transform_input` in the `forward()` function to un-normalize it.
"""
models = dict()@staticmethod
def build_model(align_tf=True):"""Builds the model and load pre-trained weights.If `align_tf` is set as True, the model will predict 1008 classes, andthe pre-trained weight from `https://github.com/mseitzer/pytorch-fid`will be loaded. Otherwise, the model will predict 1000 classes, and willload the model from `torchvision`.The built model supports following arguments when forwarding:- transform_input: Whether to transform the input back to pixel range(-1, 1). Please disable this argument if your input is already withpixel range (-1, 1). (default: False)- output_logits: Whether to output the categorical logits instead offeatures. (default: False)- remove_logits_bias: Whether to remove the bias when computing thelogits. The official implementation removes the bias by default.Please refer to`https://github.com/openai/improved-gan/blob/master/inception_score/model.py`.(default: False)- output_predictions: Whether to output the final predictions, i.e.,`softmax(logits)`. (default: False)"""if align_tf:num_classes = 1008model_source = 'tf_inception_v3'else:num_classes = 1000model_source = 'torchvision_official'fingerprint = model_sourceif fingerprint not in InceptionModel.models:# Build model.model = Inception3(num_classes=num_classes,aux_logits=False,init_weights=False,align_tf=align_tf)# Download pre-trained weights.if dist.is_initialized() and dist.get_rank() != 0:dist.barrier() # Download by chief.url, sha256 = _MODEL_URL_SHA256[model_source]filename = f'inception_model_{model_source}_{sha256}.pth'model_path, hash_check = download_url(url,filename=filename,sha256=sha256)state_dict = torch.load(model_path, map_location='cpu')if hash_check is False:warnings.warn(f'Hash check failed! The remote file from URL 'f'`{url}` may be changed, or the downloading is 'f'interrupted. The loaded inception model may 'f'have unexpected behavior.')if dist.is_initialized() and dist.get_rank() == 0:dist.barrier() # Wait for other replicas.# Load weights.model.load_state_dict(state_dict, strict=False)del state_dict# For inference only.model.eval().requires_grad_(False).cuda()InceptionModel.models[fingerprint] = modelreturn InceptionModel.models[fingerprint]
class Inception3(nn.Module):
def __init__(self, num_classes=1000, aux_logits=True, inception_blocks=None,init_weights=True, align_tf=True):super(Inception3, self).__init__()if inception_blocks is None:inception_blocks = [BasicConv2d, InceptionA, InceptionB, InceptionC,InceptionD, InceptionE, InceptionAux]assert len(inception_blocks) == 7conv_block = inception_blocks[0]inception_a = inception_blocks[1]inception_b = inception_blocks[2]inception_c = inception_blocks[3]inception_d = inception_blocks[4]inception_e = inception_blocks[5]inception_aux = inception_blocks[6]self.aux_logits = aux_logitsself.align_tf = align_tfself.Conv2d_1a_3x3 = conv_block(3, 32, kernel_size=3, stride=2)self.Conv2d_2a_3x3 = conv_block(32, 32, kernel_size=3)self.Conv2d_2b_3x3 = conv_block(32, 64, kernel_size=3, padding=1)self.Conv2d_3b_1x1 = conv_block(64, 80, kernel_size=1)self.Conv2d_4a_3x3 = conv_block(80, 192, kernel_size=3)self.Mixed_5b = inception_a(192, pool_features=32, align_tf=self.align_tf)self.Mixed_5c = inception_a(256, pool_features=64, align_tf=self.align_tf)self.Mixed_5d = inception_a(288, pool_features=64, align_tf=self.align_tf)self.Mixed_6a = inception_b(288)self.Mixed_6b = inception_c(768, channels_7x7=128, align_tf=self.align_tf)self.Mixed_6c = inception_c(768, channels_7x7=160, align_tf=self.align_tf)self.Mixed_6d = inception_c(768, channels_7x7=160, align_tf=self.align_tf)self.Mixed_6e = inception_c(768, channels_7x7=192, align_tf=self.align_tf)if aux_logits:self.AuxLogits = inception_aux(768, num_classes)self.Mixed_7a = inception_d(768)self.Mixed_7b = inception_e(1280, align_tf=self.align_tf)self.Mixed_7c = inception_e(2048, use_max_pool=self.align_tf)self.fc = nn.Linear(2048, num_classes)if init_weights:for m in self.modules():if isinstance(m, nn.Conv2d) or isinstance(m, nn.Linear):import scipy.stats as statsstddev = m.stddev if hasattr(m, 'stddev') else 0.1X = stats.truncnorm(-2, 2, scale=stddev)values = torch.as_tensor(X.rvs(m.weight.numel()), dtype=m.weight.dtype)values = values.view(m.weight.size())with torch.no_grad():m.weight.copy_(values)elif isinstance(m, nn.BatchNorm2d):nn.init.constant_(m.weight, 1)nn.init.constant_(m.bias, 0)@staticmethod
def _transform_input(x, transform_input=False):if transform_input:x_ch0 = torch.unsqueeze(x[:, 0], 1) * (0.229 / 0.5) + (0.485 - 0.5) / 0.5x_ch1 = torch.unsqueeze(x[:, 1], 1) * (0.224 / 0.5) + (0.456 - 0.5) / 0.5x_ch2 = torch.unsqueeze(x[:, 2], 1) * (0.225 / 0.5) + (0.406 - 0.5) / 0.5x = torch.cat((x_ch0, x_ch1, x_ch2), 1)return xdef _forward(self,x,output_logits=False,remove_logits_bias=False,output_predictions=False):# Upsample if necessary.if x.shape[2] != 299 or x.shape[3] != 299:if self.align_tf:theta = torch.eye(2, 3).to(x)theta[0, 2] += theta[0, 0] / x.shape[3] - theta[0, 0] / 299theta[1, 2] += theta[1, 1] / x.shape[2] - theta[1, 1] / 299theta = theta.unsqueeze(0).repeat(x.shape[0], 1, 1)grid = F.affine_grid(theta,size=(x.shape[0], x.shape[1], 299, 299),align_corners=False)x = F.grid_sample(x, grid,mode='bilinear',padding_mode='border',align_corners=False)else:x = F.interpolate(x, size=(299, 299), mode='bilinear', align_corners=False)if x.shape[1] == 1:x = x.repeat((1, 3, 1, 1))if self.align_tf:x = (x * 127.5 + 127.5 - 128) / 128# N x 3 x 299 x 299x = self.Conv2d_1a_3x3(x)# N x 32 x 149 x 149x = self.Conv2d_2a_3x3(x)# N x 32 x 147 x 147x = self.Conv2d_2b_3x3(x)# N x 64 x 147 x 147x = F.max_pool2d(x, kernel_size=3, stride=2)# N x 64 x 73 x 73x = self.Conv2d_3b_1x1(x)# N x 80 x 73 x 73x = self.Conv2d_4a_3x3(x)# N x 192 x 71 x 71x = F.max_pool2d(x, kernel_size=3, stride=2)# N x 192 x 35 x 35x = self.Mixed_5b(x)# N x 256 x 35 x 35x = self.Mixed_5c(x)# N x 288 x 35 x 35x = self.Mixed_5d(x)# N x 288 x 35 x 35x = self.Mixed_6a(x)# N x 768 x 17 x 17x = self.Mixed_6b(x)# N x 768 x 17 x 17x = self.Mixed_6c(x)# N x 768 x 17 x 17x = self.Mixed_6d(x)# N x 768 x 17 x 17x = self.Mixed_6e(x)# N x 768 x 17 x 17if self.training and self.aux_logits:aux = self.AuxLogits(x)else:aux = None# N x 768 x 17 x 17x = self.Mixed_7a(x)# N x 1280 x 8 x 8x = self.Mixed_7b(x)# N x 2048 x 8 x 8x = self.Mixed_7c(x)# N x 2048 x 8 x 8# Adaptive average poolingx = F.adaptive_avg_pool2d(x, (1, 1))# N x 2048 x 1 x 1x = F.dropout(x, training=self.training)# N x 2048 x 1 x 1x = torch.flatten(x, 1)# N x 2048if output_logits or output_predictions:x = self.fc(x)# N x 1000 (num_classes)if remove_logits_bias:x = x - self.fc.bias.view(1, -1)if output_predictions:x = F.softmax(x, dim=1)return x, auxdef forward(self,x,transform_input=False,output_logits=False,remove_logits_bias=False,output_predictions=False):x = self._transform_input(x, transform_input)x, aux = self._forward(x, output_logits, remove_logits_bias, output_predictions)if self.training and self.aux_logits:return x, auxelse:return x
class InceptionA(nn.Module):
def __init__(self, in_channels, pool_features, conv_block=None, align_tf=False):super(InceptionA, self).__init__()if conv_block is None:conv_block = BasicConv2dself.branch1x1 = conv_block(in_channels, 64, kernel_size=1)self.branch5x5_1 = conv_block(in_channels, 48, kernel_size=1)self.branch5x5_2 = conv_block(48, 64, kernel_size=5, padding=2)self.branch3x3dbl_1 = conv_block(in_channels, 64, kernel_size=1)self.branch3x3dbl_2 = conv_block(64, 96, kernel_size=3, padding=1)self.branch3x3dbl_3 = conv_block(96, 96, kernel_size=3, padding=1)self.branch_pool = conv_block(in_channels, pool_features, kernel_size=1)self.pool_include_padding = not align_tfdef _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)branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1,count_include_pad=self.pool_include_padding)branch_pool = self.branch_pool(branch_pool)outputs = [branch1x1, branch5x5, branch3x3dbl, branch_pool]return outputsdef forward(self, x):outputs = self._forward(x)return torch.cat(outputs, 1)
class InceptionB(nn.Module):
def __init__(self, in_channels, conv_block=None):super(InceptionB, self).__init__()if conv_block is None:conv_block = BasicConv2dself.branch3x3 = conv_block(in_channels, 384, kernel_size=3, stride=2)self.branch3x3dbl_1 = conv_block(in_channels, 64, kernel_size=1)self.branch3x3dbl_2 = conv_block(64, 96, kernel_size=3, padding=1)self.branch3x3dbl_3 = conv_block(96, 96, kernel_size=3, stride=2)def _forward(self, x):branch3x3 = self.branch3x3(x)branch3x3dbl = self.branch3x3dbl_1(x)branch3x3dbl = self.branch3x3dbl_2(branch3x3dbl)branch3x3dbl = self.branch3x3dbl_3(branch3x3dbl)branch_pool = F.max_pool2d(x, kernel_size=3, stride=2)outputs = [branch3x3, branch3x3dbl, branch_pool]return outputsdef forward(self, x):outputs = self._forward(x)return torch.cat(outputs, 1)
class InceptionC(nn.Module):
def __init__(self, in_channels, channels_7x7, conv_block=None, align_tf=False):super(InceptionC, self).__init__()if conv_block is None:conv_block = BasicConv2dself.branch1x1 = conv_block(in_channels, 192, kernel_size=1)c7 = channels_7x7self.branch7x7_1 = conv_block(in_channels, c7, kernel_size=1)self.branch7x7_2 = conv_block(c7, c7, kernel_size=(1, 7), padding=(0, 3))self.branch7x7_3 = conv_block(c7, 192, kernel_size=(7, 1), padding=(3, 0))self.branch7x7dbl_1 = conv_block(in_channels, c7, kernel_size=1)self.branch7x7dbl_2 = conv_block(c7, c7, kernel_size=(7, 1), padding=(3, 0))self.branch7x7dbl_3 = conv_block(c7, c7, kernel_size=(1, 7), padding=(0, 3))self.branch7x7dbl_4 = conv_block(c7, c7, kernel_size=(7, 1), padding=(3, 0))self.branch7x7dbl_5 = conv_block(c7, 192, kernel_size=(1, 7), padding=(0, 3))self.branch_pool = conv_block(in_channels, 192, kernel_size=1)self.pool_include_padding = not align_tfdef _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)branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1,count_include_pad=self.pool_include_padding)branch_pool = self.branch_pool(branch_pool)outputs = [branch1x1, branch7x7, branch7x7dbl, branch_pool]return outputsdef forward(self, x):outputs = self._forward(x)return torch.cat(outputs, 1)
class InceptionD(nn.Module):
def __init__(self, in_channels, conv_block=None):super(InceptionD, self).__init__()if conv_block is None:conv_block = BasicConv2dself.branch3x3_1 = conv_block(in_channels, 192, kernel_size=1)self.branch3x3_2 = conv_block(192, 320, kernel_size=3, stride=2)self.branch7x7x3_1 = conv_block(in_channels, 192, kernel_size=1)self.branch7x7x3_2 = conv_block(192, 192, kernel_size=(1, 7), padding=(0, 3))self.branch7x7x3_3 = conv_block(192, 192, kernel_size=(7, 1), padding=(3, 0))self.branch7x7x3_4 = conv_block(192, 192, kernel_size=3, stride=2)def _forward(self, x):branch3x3 = self.branch3x3_1(x)branch3x3 = self.branch3x3_2(branch3x3)branch7x7x3 = self.branch7x7x3_1(x)branch7x7x3 = self.branch7x7x3_2(branch7x7x3)branch7x7x3 = self.branch7x7x3_3(branch7x7x3)branch7x7x3 = self.branch7x7x3_4(branch7x7x3)branch_pool = F.max_pool2d(x, kernel_size=3, stride=2)outputs = [branch3x3, branch7x7x3, branch_pool]return outputsdef forward(self, x):outputs = self._forward(x)return torch.cat(outputs, 1)
class InceptionE(nn.Module):
def __init__(self, in_channels, conv_block=None, align_tf=False, use_max_pool=False):super(InceptionE, self).__init__()if conv_block is None:conv_block = BasicConv2dself.branch1x1 = conv_block(in_channels, 320, kernel_size=1)self.branch3x3_1 = conv_block(in_channels, 384, kernel_size=1)self.branch3x3_2a = conv_block(384, 384, kernel_size=(1, 3), padding=(0, 1))self.branch3x3_2b = conv_block(384, 384, kernel_size=(3, 1), padding=(1, 0))self.branch3x3dbl_1 = conv_block(in_channels, 448, kernel_size=1)self.branch3x3dbl_2 = conv_block(448, 384, kernel_size=3, padding=1)self.branch3x3dbl_3a = conv_block(384, 384, kernel_size=(1, 3), padding=(0, 1))self.branch3x3dbl_3b = conv_block(384, 384, kernel_size=(3, 1), padding=(1, 0))self.branch_pool = conv_block(in_channels, 192, kernel_size=1)self.pool_include_padding = not align_tfself.use_max_pool = use_max_pooldef _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)if self.use_max_pool:branch_pool = F.max_pool2d(x, kernel_size=3, stride=1, padding=1)else:branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1,count_include_pad=self.pool_include_padding)branch_pool = self.branch_pool(branch_pool)outputs = [branch1x1, branch3x3, branch3x3dbl, branch_pool]return outputsdef forward(self, x):outputs = self._forward(x)return torch.cat(outputs, 1)
class InceptionAux(nn.Module):
def __init__(self, in_channels, num_classes, conv_block=None):super(InceptionAux, self).__init__()if conv_block is None:conv_block = BasicConv2dself.conv0 = conv_block(in_channels, 128, kernel_size=1)self.conv1 = conv_block(128, 768, kernel_size=5)self.conv1.stddev = 0.01self.fc = nn.Linear(768, num_classes)self.fc.stddev = 0.001def forward(self, x):# N x 768 x 17 x 17x = F.avg_pool2d(x, kernel_size=5, stride=3)# N x 768 x 5 x 5x = self.conv0(x)# N x 128 x 5 x 5x = self.conv1(x)# N x 768 x 1 x 1# Adaptive average poolingx = F.adaptive_avg_pool2d(x, (1, 1))# N x 768 x 1 x 1x = torch.flatten(x, 1)# N x 768x = self.fc(x)# N x 1000return x
class BasicConv2d(nn.Module):
def __init__(self, in_channels, out_channels, **kwargs):super(BasicConv2d, self).__init__()self.conv = nn.Conv2d(in_channels, out_channels, bias=False, **kwargs)self.bn = nn.BatchNorm2d(out_channels, eps=0.001)def forward(self, x):x = self.conv(x)x = self.bn(x)return F.relu(x, inplace=True