1.搭建GoogleNet网络
# -*-coding:utf-8-*-
import torch.nn as nn
import torch
import torch.nn.functional as F
class GoogLeNet(nn.Module):
def __init__(self, num_classes=1000, aux_logits=True, init_weights=False):
super(GoogLeNet, self).__init__()
self.aux_logits = aux_logits
self.conv1 = BasicConv2d(3, 64, kernel_size=7, stride=2, padding=3)
self.maxpool1 = nn.MaxPool2d(3, stride=2, ceil_mode=True)
self.conv2 = BasicConv2d(64, 64, kernel_size=1)
self.conv3 = BasicConv2d(64, 192, kernel_size=3, padding=1)
self.maxpool2 = nn.MaxPool2d(3, stride=2, ceil_mode=True)
self.inception3a = Inception(192, 64, 96, 128, 16, 32, 32)
self.inception3b = Inception(256, 128, 128, 192, 32, 96, 64)
self.maxpool3 = nn.MaxPool2d(3, stride=2, ceil_mode=True)
self.inception4a = Inception(480, 192, 96, 208, 16, 48, 64)
self.inception4b = Inception(512, 160, 112, 224, 24, 64, 64)
self.inception4c = Inception(512, 128, 128, 256, 24, 64, 64)
self.inception4d = Inception(512, 112, 144, 288, 32, 64, 64)
self.inception4e = Inception(528, 256, 160, 320, 32, 128, 128)
self.maxpool4 = nn.MaxPool2d(3, stride=2, ceil_mode=True)
self.inception5a = Inception(832, 256, 160, 320, 32, 128, 128)
self.inception5b = Inception(832, 384, 192, 384, 48, 128, 128)
if self.aux_logits:
self.aux1 = InceptionAux(512, num_classes)
self.aux2 = InceptionAux(528, num_classes)
self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
self.dropout = nn.Dropout(0.4)
self.fc = nn.Linear(1024, num_classes)
if init_weights:
self._initialize_weights()
def forward(self, x):
# N x 3 x 224 x 224
x = self.conv1(x)
# N x 64 x 112 x 112
x = self.maxpool1(x)
# N x 64 x 56 x 56
x = self.conv2(x)
# N x 64 x 56 x 56
x = self.conv3(x)
# N x 192 x 56 x 56
x = self.maxpool2(x)
# N x 192 x 28 x 28
x = self.inception3a(x)
# N x 256 x 28 x 28
x = self.inception3b(x)
# N x 480 x 28 x 28
x = self.maxpool3(x)
# N x 480 x 14 x 14
x = self.inception4a(x)
# N x 512 x 14 x 14
if self.training and self.aux_logits: # eval model lose this layer
aux1 = self.aux1(x)
x = self.inception4b(x)
# N x 512 x 14 x 14
x = self.inception4c(x)
# N x 512 x 14 x 14
x = self.inception4d(x)
# N x 528 x 14 x 14
if self.training and self.aux_logits: # eval model lose this layer
aux2 = self.aux2(x)
x = self.inception4e(x)
# N x 832 x 14 x 14
x = self.maxpool4(x)
# N x 832 x 7 x 7
x = self.inception5a(x)
# N x 832 x 7 x 7
x = self.inception5b(x)
# N x 1024 x 7 x 7
x = self.avgpool(x)
# N x 1024 x 1 x 1
x = torch.flatten(x, 1)
# N x 1024
x = self.dropout(x)
x = self.fc(x)
# N x 1000 (num_classes)
if self.training and self.aux_logits: # eval model lose this layer
return x, aux2, aux1
return x
def _initialize_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
if m.bias is not None:
nn.init.constant_(m.bias, 0)
elif isinstance(m, nn.Linear):
nn.init.normal_(m.weight, 0, 0.01)
nn.init.constant_(m.bias, 0)
class Inception(nn.Module):
def __init__(self, in_channels, ch1x1, ch3x3red, ch3x3, ch5x5red, ch5x5, pool_proj):
super(Inception, self).__init__()
self.branch1 = BasicConv2d(in_channels, ch1x1, kernel_size=1)
self.branch2 = nn.Sequential(
BasicConv2d(in_channels, ch3x3red, kernel_size=1),
BasicConv2d(ch3x3red, ch3x3, kernel_size=3, padding=1) # 保证输出大小等于输入大小
)
self.branch3 = nn.Sequential(
BasicConv2d(in_channels, ch5x5red, kernel_size=1),
BasicConv2d(ch5x5red, ch5x5, kernel_size=5, padding=2) # 保证输出大小等于输入大小
)
self.branch4 = nn.Sequential(
nn.MaxPool2d(kernel_size=3, stride=1, padding=1),
BasicConv2d(in_channels, pool_proj, kernel_size=1)
)
def forward(self, x):
branch1 = self.branch1(x)
branch2 = self.branch2(x)
branch3 = self.branch3(x)
branch4 = self.branch4(x)
outputs = [branch1, branch2, branch3, branch4]
return torch.cat(outputs, 1)
class InceptionAux(nn.Module):
def __init__(self, in_channels, num_classes):
super(InceptionAux, self).__init__()
self.averagePool = nn.AvgPool2d(kernel_size=5, stride=3)
self.conv = BasicConv2d(in_channels, 128, kernel_size=1) # output[batch, 128, 4, 4]
self.fc1 = nn.Linear(2048, 1024)
self.fc2 = nn.Linear(1024, num_classes)
def forward(self, x):
# aux1: N x 512 x 14 x 14, aux2: N x 528 x 14 x 14
x = self.averagePool(x)
# aux1: N x 512 x 4 x 4, aux2: N x 528 x 4 x 4
x = self.conv(x)
# N x 128 x 4 x 4
x = torch.flatten(x, 1)
x = F.dropout(x, 0.5, training=self.training)
# N x 2048
x = F.relu(self.fc1(x), inplace=True)
x = F.dropout(x, 0.5, training=self.training)
# N x 1024
x = self.fc2(x)
# N x num_classes
return 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, **kwargs)
self.relu = nn.ReLU(inplace=True)
def forward(self, x):
x = self.conv(x)
x = self.relu(x)
return x
2.训练网络
# -*-coding:utf-8-*-
import os
import json
import torch
import torch.nn as nn
from torchvision import transforms, datasets
import torch.optim as optim
from tqdm import tqdm
from Googlenet_model import GoogLeNet
def main():
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print("using {} device.".format(device))
data_transform = {
"train": transforms.Compose([transforms.RandomResizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))]),
"val": transforms.Compose([transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])}
data_root = os.path.abspath(os.path.join(os.getcwd(), "../")) # get data root path
image_path = os.path.join(data_root, "data_set") # flower data set path
assert os.path.exists(image_path), "{} path does not exist.".format(image_path)
train_dataset = datasets.ImageFolder(root=os.path.join(image_path, "train"),
transform=data_transform["train"])
train_num = len(train_dataset)
# {'daisy':0, 'dandelion':1, 'roses':2, 'sunflower':3, 'tulips':4}
flower_list = train_dataset.class_to_idx
cla_dict = dict((val, key) for key, val in flower_list.items())
# write dict into json file
json_str = json.dumps(cla_dict, indent=4)
with open('class_indices.json', 'w') as json_file:
json_file.write(json_str)
batch_size = 5
nw = min([os.cpu_count(), batch_size if batch_size > 1 else 0, 8]) # number of workers
print('Using {} dataloader workers every process'.format(nw))
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=batch_size, shuffle=True,
num_workers=nw)
validate_dataset = datasets.ImageFolder(root=os.path.join(image_path, "val"),
transform=data_transform["val"])
val_num = len(validate_dataset)
validate_loader = torch.utils.data.DataLoader(validate_dataset,
batch_size=batch_size, shuffle=False,
num_workers=nw)
print("using {} images for training, {} images for validation.".format(train_num,
val_num))
# test_data_iter = iter(validate_loader)
# test_image, test_label = test_data_iter.next()
# net = torchvision.models.googlenet(num_classes=5)
# model_dict = net.state_dict()
# pretrain_model = torch.load("googlenet.pth")
# del_list = ["aux1.fc2.weight", "aux1.fc2.bias",
# "aux2.fc2.weight", "aux2.fc2.bias",
# "fc.weight", "fc.bias"]
# pretrain_dict = {k: v for k, v in pretrain_model.items() if k not in del_list}
# model_dict.update(pretrain_dict)
# net.load_state_dict(model_dict)
net = GoogLeNet(num_classes=7, aux_logits=True, init_weights=True)
net.to(device)
loss_function = nn.CrossEntropyLoss()
optimizer = optim.Adam(net.parameters(), lr=0.0003)
epochs = 30
best_acc = 0.0
save_path = './googleNet.pth'
train_steps = len(train_loader)
for epoch in range(epochs):
# train
net.train()
running_loss = 0.0
train_bar = tqdm(train_loader)
for step, data in enumerate(train_bar):
images, labels = data
optimizer.zero_grad()
logits, aux_logits2, aux_logits1 = net(images.to(device))
loss0 = loss_function(logits, labels.to(device))
loss1 = loss_function(aux_logits1, labels.to(device))
loss2 = loss_function(aux_logits2, labels.to(device))
loss = loss0 + loss1 * 0.3 + loss2 * 0.3
loss.backward()
optimizer.step()
# print statistics
running_loss += loss.item()
train_bar.desc = "train epoch[{}/{}] loss:{:.3f}".format(epoch + 1,
epochs,
loss)
# validate
net.eval()
acc = 0.0 # accumulate accurate number / epoch
with torch.no_grad():
val_bar = tqdm(validate_loader)
for val_data in val_bar:
val_images, val_labels = val_data
outputs = net(val_images.to(device)) # eval model only have last output layer
predict_y = torch.max(outputs, dim=1)[1]
acc += torch.eq(predict_y, val_labels.to(device)).sum().item()
val_accurate = acc / val_num
print('[epoch %d] train_loss: %.3f val_accuracy: %.3f' %
(epoch + 1, running_loss / train_steps, val_accurate))
if val_accurate > best_acc:
best_acc = val_accurate
torch.save(net.state_dict(), save_path)
print('Finished Training')
if __name__ == '__main__':
main()
3.训练结果
/home/xyjin/anaconda3/envs/pytorch/bin/python /home/xyjin/PycharmProjects/data_mining/data_process/rock/classed_data/Googlenet/Googlenet_train.py
using cpu device.
Using 4 dataloader workers every process
using 12600 images for training, 1400 images for validation.
train epoch[1/30] loss:2.236: 100%|██████████| 2520/2520 [31:44<00:00, 1.32it/s]
100%|██████████| 280/280 [01:29<00:00, 3.11it/s]
[epoch 1] train_loss: 2.098 val_accuracy: 0.646
train epoch[2/30] loss:1.399: 100%|██████████| 2520/2520 [31:24<00:00, 1.34it/s]
100%|██████████| 280/280 [01:28<00:00, 3.16it/s]
[epoch 2] train_loss: 1.585 val_accuracy: 0.679
train epoch[3/30] loss:0.996: 100%|██████████| 2520/2520 [31:22<00:00, 1.34it/s]
100%|██████████| 280/280 [01:26<00:00, 3.24it/s]
[epoch 3] train_loss: 1.378 val_accuracy: 0.761
train epoch[4/30] loss:0.939: 100%|██████████| 2520/2520 [31:25<00:00, 1.34it/s]
100%|██████████| 280/280 [01:26<00:00, 3.24it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 4] train_loss: 1.206 val_accuracy: 0.750
train epoch[5/30] loss:0.992: 100%|██████████| 2520/2520 [31:26<00:00, 1.34it/s]
100%|██████████| 280/280 [01:26<00:00, 3.23it/s]
[epoch 5] train_loss: 1.087 val_accuracy: 0.813
train epoch[6/30] loss:0.262: 100%|██████████| 2520/2520 [31:39<00:00, 1.33it/s]
100%|██████████| 280/280 [01:30<00:00, 3.09it/s]
[epoch 6] train_loss: 0.970 val_accuracy: 0.823
train epoch[7/30] loss:0.754: 100%|██████████| 2520/2520 [31:30<00:00, 1.33it/s]
100%|██████████| 280/280 [01:25<00:00, 3.27it/s]
[epoch 7] train_loss: 0.875 val_accuracy: 0.848
train epoch[8/30] loss:0.957: 100%|██████████| 2520/2520 [31:25<00:00, 1.34it/s]
100%|██████████| 280/280 [01:28<00:00, 3.17it/s]
[epoch 8] train_loss: 0.793 val_accuracy: 0.861
train epoch[9/30] loss:0.269: 100%|██████████| 2520/2520 [31:27<00:00, 1.34it/s]
100%|██████████| 280/280 [01:25<00:00, 3.27it/s]
[epoch 9] train_loss: 0.745 val_accuracy: 0.916
train epoch[10/30] loss:0.574: 100%|██████████| 2520/2520 [31:26<00:00, 1.34it/s]
100%|██████████| 280/280 [01:25<00:00, 3.27it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 10] train_loss: 0.700 val_accuracy: 0.904
train epoch[11/30] loss:0.876: 100%|██████████| 2520/2520 [31:24<00:00, 1.34it/s]
100%|██████████| 280/280 [01:28<00:00, 3.18it/s]
[epoch 11] train_loss: 0.664 val_accuracy: 0.901
train epoch[12/30] loss:0.196: 100%|██████████| 2520/2520 [31:21<00:00, 1.34it/s]
100%|██████████| 280/280 [01:28<00:00, 3.17it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 12] train_loss: 0.636 val_accuracy: 0.910
train epoch[13/30] loss:0.360: 100%|██████████| 2520/2520 [31:37<00:00, 1.33it/s]
100%|██████████| 280/280 [01:24<00:00, 3.29it/s]
[epoch 13] train_loss: 0.625 val_accuracy: 0.934
train epoch[14/30] loss:0.990: 100%|██████████| 2520/2520 [31:27<00:00, 1.34it/s]
100%|██████████| 280/280 [01:30<00:00, 3.09it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 14] train_loss: 0.576 val_accuracy: 0.916
train epoch[15/30] loss:0.396: 100%|██████████| 2520/2520 [31:31<00:00, 1.33it/s]
100%|██████████| 280/280 [01:24<00:00, 3.30it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 15] train_loss: 0.548 val_accuracy: 0.905
train epoch[16/30] loss:0.073: 100%|██████████| 2520/2520 [31:23<00:00, 1.34it/s]
100%|██████████| 280/280 [01:25<00:00, 3.29it/s]
[epoch 16] train_loss: 0.538 val_accuracy: 0.929
train epoch[17/30] loss:1.332: 100%|██████████| 2520/2520 [31:33<00:00, 1.33it/s]
100%|██████████| 280/280 [01:25<00:00, 3.28it/s]
[epoch 17] train_loss: 0.510 val_accuracy: 0.959
train epoch[18/30] loss:2.095: 100%|██████████| 2520/2520 [31:33<00:00, 1.33it/s]
100%|██████████| 280/280 [01:26<00:00, 3.25it/s]
[epoch 18] train_loss: 0.493 val_accuracy: 0.954
train epoch[19/30] loss:1.133: 100%|██████████| 2520/2520 [31:34<00:00, 1.33it/s]
100%|██████████| 280/280 [01:28<00:00, 3.18it/s]
[epoch 19] train_loss: 0.495 val_accuracy: 0.888
train epoch[20/30] loss:0.198: 100%|██████████| 2520/2520 [31:41<00:00, 1.32it/s]
100%|██████████| 280/280 [01:28<00:00, 3.17it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 20] train_loss: 0.496 val_accuracy: 0.925
train epoch[21/30] loss:0.703: 100%|██████████| 2520/2520 [31:41<00:00, 1.33it/s]
100%|██████████| 280/280 [01:28<00:00, 3.17it/s]
[epoch 21] train_loss: 0.474 val_accuracy: 0.910
train epoch[22/30] loss:0.508: 100%|██████████| 2520/2520 [31:38<00:00, 1.33it/s]
100%|██████████| 280/280 [01:27<00:00, 3.22it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 22] train_loss: 0.454 val_accuracy: 0.946
train epoch[23/30] loss:0.068: 100%|██████████| 2520/2520 [31:46<00:00, 1.32it/s]
100%|██████████| 280/280 [01:24<00:00, 3.30it/s]
[epoch 23] train_loss: 0.445 val_accuracy: 0.972
train epoch[24/30] loss:0.089: 100%|██████████| 2520/2520 [31:53<00:00, 1.32it/s]
100%|██████████| 280/280 [01:24<00:00, 3.31it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 24] train_loss: 0.450 val_accuracy: 0.937
train epoch[25/30] loss:0.041: 100%|██████████| 2520/2520 [33:04<00:00, 1.27it/s]
100%|██████████| 280/280 [01:27<00:00, 3.20it/s]
[epoch 25] train_loss: 0.419 val_accuracy: 0.956
train epoch[26/30] loss:0.296: 100%|██████████| 2520/2520 [32:28<00:00, 1.29it/s]
100%|██████████| 280/280 [01:28<00:00, 3.17it/s]
[epoch 26] train_loss: 0.406 val_accuracy: 0.951
train epoch[27/30] loss:0.318: 100%|██████████| 2520/2520 [32:36<00:00, 1.29it/s]
100%|██████████| 280/280 [01:25<00:00, 3.27it/s]
0%| | 0/2520 [00:00<?, ?it/s][epoch 27] train_loss: 0.423 val_accuracy: 0.965
train epoch[28/30] loss:0.197: 100%|██████████| 2520/2520 [32:37<00:00, 1.29it/s]
100%|██████████| 280/280 [01:28<00:00, 3.16it/s]
[epoch 28] train_loss: 0.402 val_accuracy: 0.970
train epoch[29/30] loss:0.001: 100%|██████████| 2520/2520 [32:42<00:00, 1.28it/s]
100%|██████████| 280/280 [01:29<00:00, 3.12it/s]
[epoch 29] train_loss: 0.413 val_accuracy: 0.948
train epoch[30/30] loss:0.847: 100%|██████████| 2520/2520 [32:51<00:00, 1.28it/s]
100%|██████████| 280/280 [01:24<00:00, 3.31it/s]
[epoch 30] train_loss: 0.392 val_accuracy: 0.964
Finished Training
Process finished with exit code 0