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 torch.backends.cudnn as cudnn
import numpy as np
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import os
import copy
cudnn.benchmark = True
plt.ion() # interactive mode
transform = transforms.Compose(
[transforms.ToTensor()]
)
trainset = torchvision.datasets.MNIST(root='./data', train=True,
download=True, transform=transform)
testset = torchvision.datasets.MNIST(root='./data', train=False,
download=True, transform=transform)
dataloaders = {'train':torch.utils.data.DataLoader(trainset, batch_size=batch_size,
shuffle=True, num_workers=2),
'val':torch.utils.data.DataLoader(testset, batch_size=batch_size,
shuffle=False, num_workers=2)
}
dataset_sizes = {'train':60000, 'val':10000}
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(f'Epoch {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(f'{phase} Loss: {epoch_loss:.4f} Acc: {epoch_acc:.4f}')
# 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(f'Training complete in {time_elapsed // 60:.0f}m {time_elapsed % 60:.0f}s')
print(f'Best val Acc: {best_acc:4f}')
# load best model weights
model.load_state_dict(best_model_wts)
return model
import torch.nn as nn
import torch.nn.functional as F
class NeuralNetwork(nn.Module):
def __init__(self):
super(NeuralNetwork, self).__init__()
self.linear_relu_stack = nn.Sequential(
nn.Conv2d(1, 16, 3),
nn.ReLU(),
nn.MaxPool2d(2, 2),
nn.Conv2d(16, 32, 3),
nn.ReLU(),
nn.MaxPool2d(2, 2),
nn.Conv2d(32, 64, 3),
nn.ReLU(),
nn.Flatten(),
nn.Linear(64*3*3, 64),
nn.ReLU(),
nn.Linear(64, 10)
)
def forward(self, x):
logits = self.linear_relu_stack(x)
return logits
model_ft = NeuralNetwork()
print(model_ft)
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model_ft = model_ft.to(device)
criterion = nn.CrossEntropyLoss()
# Observe that all parameters are being optimized
optimizer_ft = optim.Adam(model_ft.parameters())
# 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)
NeuralNetwork(
(linear_relu_stack): Sequential(
(0): Conv2d(1, 16, kernel_size=(3, 3), stride=(1, 1))
(1): ReLU()
(2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(3): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1))
(4): ReLU()
(5): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(6): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1))
(7): ReLU()
(8): Flatten(start_dim=1, end_dim=-1)
(9): Linear(in_features=576, out_features=64, bias=True)
(10): ReLU()
(11): Linear(in_features=64, out_features=10, bias=True)
)
)
Epoch 0/9
----------
train Loss: 0.1799 Acc: 0.9451
val Loss: 0.0493 Acc: 0.9834
Epoch 1/9
----------
train Loss: 0.0566 Acc: 0.9819
val Loss: 0.0512 Acc: 0.9838
Epoch 2/9
----------
train Loss: 0.0380 Acc: 0.9882
val Loss: 0.0618 Acc: 0.9804
Epoch 3/9
----------
train Loss: 0.0307 Acc: 0.9903
val Loss: 0.0398 Acc: 0.9875
Epoch 4/9
----------
train Loss: 0.0238 Acc: 0.9925
val Loss: 0.0351 Acc: 0.9886
Epoch 5/9
----------
train Loss: 0.0201 Acc: 0.9936
val Loss: 0.0416 Acc: 0.9878
Epoch 6/9
----------
train Loss: 0.0173 Acc: 0.9945
val Loss: 0.0268 Acc: 0.9907
Epoch 7/9
----------
train Loss: 0.0056 Acc: 0.9985
val Loss: 0.0256 Acc: 0.9917
Epoch 8/9
----------
train Loss: 0.0035 Acc: 0.9991
val Loss: 0.0253 Acc: 0.9927
Epoch 9/9
----------
train Loss: 0.0025 Acc: 0.9994
val Loss: 0.0262 Acc: 0.9924
Training complete in 4m 40s
Best val Acc: 0.992700
