I have trained a model using the following code in test_custom_resnet18.ipynb.
from __future__ import print_function, division
import torch
import torch.nn as nn
import torch.optim as optim
from torch.optim import lr_scheduler
import numpy as np
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import os
import copy
plt.ion()
# Data augmentation and normalization for training
# Just normalization for validation
data_transforms = {
'train': transforms.Compose([
transforms.RandomResizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
]),
'val': transforms.Compose([
transforms.Resize(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
]),
}
data_dir = 'hymenoptera_data'
image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x),
data_transforms[x])
for x in ['train', 'val']}
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=4,
shuffle=True, num_workers=4)
for x in ['train', 'val']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'val']}
class_names = image_datasets['train'].classes
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
def imshow(inp, title=None):
"""Imshow for Tensor."""
inp = inp.numpy().transpose((1, 2, 0))
mean = np.array([0.485, 0.456, 0.406])
std = np.array([0.229, 0.224, 0.225])
inp = std * inp + mean
inp = np.clip(inp, 0, 1)
plt.imshow(inp)
if title is not None:
plt.title(title)
plt.pause(0.001) # pause a bit so that plots are updated
# Get a batch of training data
inputs, classes = next(iter(dataloaders['train']))
# Make a grid from batch
out = torchvision.utils.make_grid(inputs)
imshow(out, title=[class_names[x] for x in classes])
def train_model(model, criterion, optimizer, scheduler, num_epochs=25):
since = time.time()
best_model_wts = copy.deepcopy(model.state_dict())
best_acc = 0.0
for epoch in range(num_epochs):
print('Epoch {}/{}'.format(epoch, num_epochs - 1))
print('-' * 10)
# Each epoch has a training and validation phase
for phase in ['train', 'val']:
if phase == 'train':
model.train() # Set model to training mode
else:
model.eval() # Set model to evaluate mode
running_loss = 0.0
running_corrects = 0
# Iterate over data.
for inputs, labels in dataloaders[phase]:
inputs = inputs.to(device)
labels = labels.to(device)
# zero the parameter gradients
optimizer.zero_grad()
# forward
# track history if only in train
with torch.set_grad_enabled(phase == 'train'):
outputs = model(inputs)
_, preds = torch.max(outputs, 1)
loss = criterion(outputs, labels)
# backward + optimize only if in training phase
if phase == 'train':
loss.backward()
optimizer.step()
# statistics
running_loss += loss.item() * inputs.size(0)
running_corrects += torch.sum(preds == labels.data)
if phase == 'train':
scheduler.step()
epoch_loss = running_loss / dataset_sizes[phase]
epoch_acc = running_corrects.double() / dataset_sizes[phase]
print('{} Loss: {:.4f} Acc: {:.4f}'.format(
phase, epoch_loss, epoch_acc))
# deep copy the model
if phase == 'val' and epoch_acc > best_acc:
best_acc = epoch_acc
best_model_wts = copy.deepcopy(model.state_dict())
print()
time_elapsed = time.time() - since
print('Training complete in {:.0f}m {:.0f}s'.format(
time_elapsed // 60, time_elapsed % 60))
print('Best val Acc: {:4f}'.format(best_acc))
# load best model weights
model.load_state_dict(best_model_wts)
return model
# using the custom resnet18
import custom_resnet18
model_ft = custom_resnet18.ResNet18()
num_ftrs = model_ft.fc.in_features
# Here the size of each output sample is set to 2.
# Alternatively, it can be generalized to nn.Linear(num_ftrs, len(class_names)).
model_ft.fc = nn.Linear(num_ftrs, 2)
model_ft = model_ft.to(device)
criterion = nn.CrossEntropyLoss()
# Observe that all parameters are being optimized
optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)
# Decay LR by a factor of 0.1 every 7 epochs
exp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)
model_ft = train_model(model_ft, criterion, optimizer_ft,
exp_lr_scheduler, num_epochs=25)
wherein, custom_resnet18.py is:
import torch
import torch.nn as nn
class Block(nn.Module):
def __init__(self, num_layers, in_channels, out_channels, identity_downsample=None, stride=1):
assert num_layers in [18, 34, 50, 101, 152], "should be a a valid architecture"
super(Block, self).__init__()
self.num_layers = num_layers
if self.num_layers > 34:
self.expansion = 4
else:
self.expansion = 1
# ResNet50, 101, and 152 include additional layer of 1x1 kernels
self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=1, padding=0)
self.bn1 = nn.BatchNorm2d(out_channels)
if self.num_layers > 34:
self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=stride, padding=1)
else:
# for ResNet18 and 34, connect input directly to (3x3) kernel (skip first (1x1))
self.conv2 = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=stride, padding=1)
self.bn2 = nn.BatchNorm2d(out_channels)
self.conv3 = nn.Conv2d(out_channels, out_channels * self.expansion, kernel_size=1, stride=1, padding=0)
self.bn3 = nn.BatchNorm2d(out_channels * self.expansion)
self.relu = nn.ReLU()
self.identity_downsample = identity_downsample
def forward(self, x):
identity = x
if self.num_layers > 34:
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
x = self.conv2(x)
x = self.bn2(x)
x = self.relu(x)
x = self.conv3(x)
x = self.bn3(x)
if self.identity_downsample is not None:
identity = self.identity_downsample(identity)
x += identity
x = self.relu(x)
return x
class ResNet(nn.Module):
def __init__(self, num_layers, block, image_channels, num_classes):
assert num_layers in [18, 34, 50, 101, 152], f'ResNet{num_layers}: Unknown architecture! Number of layers has ' \
f'to be 18, 34, 50, 101, or 152 '
super(ResNet, self).__init__()
if num_layers < 50:
self.expansion = 1
else:
self.expansion = 4
if num_layers == 18:
layers = [2, 2, 2, 2]
elif num_layers == 34 or num_layers == 50:
layers = [3, 4, 6, 3]
elif num_layers == 101:
layers = [3, 4, 23, 3]
else:
layers = [3, 8, 36, 3]
self.in_channels = 64
self.conv1 = nn.Conv2d(image_channels, 64, kernel_size=7, stride=2, padding=3)
self.bn1 = nn.BatchNorm2d(64)
self.relu = nn.ReLU()
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
# ResNetLayers
self.layer1 = self.make_layers(num_layers, block, layers[0], intermediate_channels=64, stride=1)
self.layer2 = self.make_layers(num_layers, block, layers[1], intermediate_channels=128, stride=2)
self.layer3 = self.make_layers(num_layers, block, layers[2], intermediate_channels=256, stride=2)
self.layer4 = self.make_layers(num_layers, block, layers[3], intermediate_channels=512, stride=2)
self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
self.fc = nn.Linear(512 * self.expansion, num_classes)
def forward(self, x):
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
x = self.maxpool(x)
x = self.layer1(x)
x = self.layer2(x)
x = self.layer3(x)
x = self.layer4(x)
x = self.avgpool(x)
x = x.reshape(x.shape[0], -1)
x = self.fc(x)
return x
def make_layers(self, num_layers, block, num_residual_blocks, intermediate_channels, stride):
layers = []
identity_downsample = nn.Sequential(nn.Conv2d(self.in_channels, intermediate_channels*self.expansion, kernel_size=1, stride=stride),
nn.BatchNorm2d(intermediate_channels*self.expansion))
layers.append(block(num_layers, self.in_channels, intermediate_channels, identity_downsample, stride))
self.in_channels = intermediate_channels * self.expansion # 256
for i in range(num_residual_blocks - 1):
layers.append(block(num_layers, self.in_channels, intermediate_channels)) # 256 -> 64, 64*4 (256) again
return nn.Sequential(*layers)
def ResNet18(img_channels=3, num_classes=1000):
return ResNet(18, Block, img_channels, num_classes)
def ResNet34(img_channels=3, num_classes=1000):
return ResNet(34, Block, img_channels, num_classes)
def ResNet50(img_channels=3, num_classes=1000):
return ResNet(50, Block, img_channels, num_classes)
def ResNet101(img_channels=3, num_classes=1000):
return ResNet(101, Block, img_channels, num_classes)
def ResNet152(img_channels=3, num_classes=1000):
return ResNet(152, Block, img_channels, num_classes)
def test():
net = ResNet18(img_channels=3, num_classes=1000)
y = net(torch.randn(4, 3, 224, 224)).to("cuda")
print(y.size())
test()
I want to extract the feature vector right before the fully-connected (FC) layer. When I use the following code, it only prints the model not the actual value of feature vector. How can I access the actual value of the feature vector?
nn.Sequential(*list(model_ft.children())[:-1])
It prints:
Sequential(
(0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU()
(3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
(4): Sequential(
(0): Block(
(conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
(identity_downsample): Sequential(
(0): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Block(
(conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
)
)
(5): Sequential(
(0): Block(
(conv1): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
(identity_downsample): Sequential(
(0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Block(
(conv1): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
)
)
(6): Sequential(
(0): Block(
(conv1): Conv2d(128, 256, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
(identity_downsample): Sequential(
(0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Block(
(conv1): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
)
)
(7): Sequential(
(0): Block(
(conv1): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
(identity_downsample): Sequential(
(0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2))
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): Block(
(conv1): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1))
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv3): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1))
(bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU()
)
)
(8): AdaptiveAvgPool2d(output_size=(1, 1))
)
I also have:
children_counter = 0
for n,c in model_ft.named_children():
print("Children Counter: ",children_counter," Layer Name: ",n,)
children_counter+=1
Output:
Children Counter: 0 Layer Name: conv1
Children Counter: 1 Layer Name: bn1
Children Counter: 2 Layer Name: relu
Children Counter: 3 Layer Name: maxpool
Children Counter: 4 Layer Name: layer1
Children Counter: 5 Layer Name: layer2
Children Counter: 6 Layer Name: layer3
Children Counter: 7 Layer Name: layer4
Children Counter: 8 Layer Name: avgpool
Children Counter: 9 Layer Name: fc
Codes are from Transfer Learning for Computer Vision Tutorial — PyTorch Tutorials 1.9.0+cu102 documentation and ResNet.ipynb · GitHub