AML21: 01 Deep Neural Network for MNIST¶
Data and Libraries¶
In [1]:
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
import matplotlib.pyplot as plt
import numpy
# this 'device' will be used for training our model
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print(device)
Load the MNIST dataset¶
Observe that we set shuffle=True, which means that data is randomized
In [2]:
input_size = 28*28 # images are 28x28 pixels
output_size = 10 # there are 10 classes
train_loader = torch.utils.data.DataLoader(
datasets.MNIST('../data', train=True, download=True,
transform=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=64, shuffle=True)
test_loader = torch.utils.data.DataLoader(
datasets.MNIST('../data', train=False, transform=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=1000, shuffle=True)
In [3]:
# show some training images
plt.figure(figsize=(16, 4))
# fetch a batch of train images; RANDOM
image_batch, label_batch = next(iter(train_loader))
for i in range(20):
image = image_batch[i]
label = label_batch[i].item()
plt.subplot(2, 10, i + 1)
#image, label = train_loader.dataset.__getitem__(i)
plt.imshow(image.squeeze().numpy())
plt.axis('off')
plt.title(label)
A 2-hidden layer Fully Connected Neural Network¶
Helper functions for training and testing¶
In [4]:
# function to count number of parameters
def get_n_params(model):
np=0
for p in list(model.parameters()):
np += p.nelement()
return np
accuracy_list = []
# we pass a model object to this trainer, and it trains this model for one epoch
def train(epoch, model):
model.train()
for batch_idx, (data, target) in enumerate(train_loader):
# send to device
data, target = data.to(device), target.to(device)
optimizer.zero_grad()
output = model(data)
loss = F.nll_loss(output, target)
loss.backward()
optimizer.step()
if batch_idx % 100 == 0:
print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
epoch, batch_idx * len(data), len(train_loader.dataset),
100. * batch_idx / len(train_loader), loss.item()))
def test(model):
model.eval()
test_loss = 0
correct = 0
for data, target in test_loader:
# send to device
data, target = data.to(device), target.to(device)
output = model(data)
test_loss += F.nll_loss(output, target, reduction='sum').item() # sum up batch loss
pred = output.data.max(1, keepdim=True)[1] # get the index of the max log-probability
correct += pred.eq(target.data.view_as(pred)).cpu().sum().item()
test_loss /= len(test_loader.dataset)
accuracy = 100. * correct / len(test_loader.dataset)
accuracy_list.append(accuracy)
print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
test_loss, correct, len(test_loader.dataset),
accuracy))
Defining the Fully Connected Network¶
In [5]:
class FC2Layer(nn.Module):
def __init__(self, input_size, output_size):
super(FC2Layer, self).__init__()
self.input_size = input_size
self.network = nn.Sequential(
nn.Linear(input_size, 200),
nn.ReLU(),
nn.Linear(200,100),
nn.ReLU(),
nn.Linear(100,60),
nn.ReLU(),
nn.Linear(60, output_size),
nn.LogSoftmax(dim=1)
)
def forward(self, x):
x = x.view(-1, self.input_size)
return self.network(x)
Train the Network¶
In [6]:
print("Training on ", device)
model_fnn = FC2Layer(input_size, output_size)
model_fnn.to(device)
optimizer = optim.SGD(model_fnn.parameters(), lr=0.01, momentum=0.5)
print('Number of parameters: {}'.format(get_n_params(model_fnn)))
for epoch in range(0, 10):
train(epoch, model_fnn)
test(model_fnn)
Show some predictions of the test network¶
In [7]:
def visualize_pred(img, pred_prob, real_label):
''' Function for viewing an image and it's predicted classes.
'''
#pred_prob = pred_prob.data.numpy().squeeze()
fig, (ax1, ax2) = plt.subplots(figsize=(6,9), ncols=2)
ax1.imshow(img.numpy().squeeze())
ax1.axis('off')
pred_label = numpy.argmax(pred_prob)
ax1.set_title([real_label, pred_label])
ax2.barh(numpy.arange(10), pred_prob)
ax2.set_aspect(0.1)
ax2.set_yticks(numpy.arange(10))
ax2.set_yticklabels(numpy.arange(10))
ax2.set_title('Prediction Probability')
ax2.set_xlim(0, 1.1)
plt.tight_layout()
In [8]:
model_fnn.to('cpu')
# fetch a batch of test images
image_batch, label_batch = next(iter(test_loader))
# Turn off gradients to speed up this part
with torch.no_grad():
log_pred_prob_batch = model_fnn(image_batch)
for i in range(10):
img = image_batch[i]
real_label = label_batch[i].item()
log_pred_prob = log_pred_prob_batch[i]
# Output of the network are log-probabilities, need to take exponential for probabilities
pred_prob = torch.exp(log_pred_prob).data.numpy().squeeze()
visualize_pred(img, pred_prob, real_label)
Network with Dropout¶
In [9]:
class FC2LayerDropout(nn.Module):
def __init__(self, input_size, output_size):
super(FC2LayerDropout, self).__init__()
self.input_size = input_size
self.network = nn.Sequential(
nn.Linear(input_size, 200),
nn.Dropout(0.25),
nn.ReLU(),
nn.Linear(200, 100),
nn.Dropout(0.25),
nn.ReLU(),
nn.Linear(100,60),
nn.ReLU(),
nn.Linear(60, output_size),
nn.LogSoftmax(dim=1)
)
def forward(self, x):
x = x.view(-1, self.input_size)
return self.network(x)
In [10]:
print("Training on ", device)
model = FC2LayerDropout(input_size, output_size)
model.to(device)
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)
print('Number of parameters: {}'.format(get_n_params(model)))
for epoch in range(0, 10):
model.train() # model in training mode. Turns on dropout, batch-norm etc during training
train(epoch, model)
model.eval() # model in evaluation mode. Turn off dropout, batch-norm etc during validation/testing
test(model)
In [11]:
model.to('cpu')
# fetch a batch of test images
image_batch, label_batch = next(iter(test_loader))
# Turn off gradients to speed up this part
with torch.no_grad():
log_pred_prob_batch = model(image_batch)
for i in range(10):
img = image_batch[i]
real_label = label_batch[i].item()
log_pred_prob = log_pred_prob_batch[i]
# Output of the network are log-probabilities, need to take exponential for probabilities
pred_prob = torch.exp(log_pred_prob).data.numpy().squeeze()
visualize_pred(img, pred_prob, real_label)
Does the Fully Connected Network use "Visual Information" ?¶
In [12]:
fixed_perm = torch.randperm(784) # Fix a permutation of the image pixels; We apply the same permutation to all images
# show some training images
plt.figure(figsize=(8, 8))
# fetch a batch of train images; RANDOM
image_batch, label_batch = next(iter(train_loader))
for i in range(6):
image = image_batch[i]
image_perm = image.view(-1, 28*28).clone()
image_perm = image_perm[:, fixed_perm]
image_perm = image_perm.view(-1, 1, 28, 28)
label = label_batch[i].item()
plt.subplot(3,4 , 2*i + 1)
#image, label = train_loader.dataset.__getitem__(i)
plt.imshow(image.squeeze().numpy())
plt.axis('off')
plt.title(label)
plt.subplot(3, 4, 2*i+2)
plt.imshow(image_perm.squeeze().numpy())
plt.axis('off')
plt.title(label)
In [13]:
accuracy_list = []
def scramble_train(epoch, model, perm=torch.arange(0, 784).long()):
model.train()
for batch_idx, (data, target) in enumerate(train_loader):
# send to device
data, target = data.to(device), target.to(device)
# permute pixels
data = data.view(-1, 28*28)
data = data[:, perm]
data = data.view(-1, 1, 28, 28)
optimizer.zero_grad()
output = model(data)
loss = F.nll_loss(output, target)
loss.backward()
optimizer.step()
if batch_idx % 100 == 0:
print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
epoch, batch_idx * len(data), len(train_loader.dataset),
100. * batch_idx / len(train_loader), loss.item()))
def scramble_test(model, perm=torch.arange(0, 784).long()):
model.eval()
test_loss = 0
correct = 0
for data, target in test_loader:
# send to device
data, target = data.to(device), target.to(device)
# permute pixels
data = data.view(-1, 28*28)
data = data[:, perm]
data = data.view(-1, 1, 28, 28)
output = model(data)
test_loss += F.nll_loss(output, target, reduction='sum').item() # sum up batch loss
pred = output.data.max(1, keepdim=True)[1] # get the index of the max log-probability
correct += pred.eq(target.data.view_as(pred)).cpu().sum().item()
test_loss /= len(test_loader.dataset)
accuracy = 100. * correct / len(test_loader.dataset)
accuracy_list.append(accuracy)
print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
test_loss, correct, len(test_loader.dataset),
accuracy))
In [14]:
print("Training on ", device)
model_fnn_2 = FC2Layer(input_size, output_size)
model_fnn_2.to(device)
optimizer = optim.SGD(model_fnn_2.parameters(), lr=0.01, momentum=0.5)
print('Number of parameters: {}'.format(get_n_params(model_fnn_2)))
for epoch in range(0, 10):
scramble_train(epoch, model_fnn_2, fixed_perm)
scramble_test(model_fnn_2, fixed_perm)
In [15]:
model_fnn_2.to('cpu')
# fetch a batch of test images
image_batch, label_batch = next(iter(test_loader))
image_batch_scramble = image_batch.view(-1, 28*28)
image_batch_scramble = image_batch_scramble[:, fixed_perm]
image_batch_scramble = image_batch_scramble.view(-1, 1, 28, 28)
# Turn off gradients to speed up this part
with torch.no_grad():
log_pred_prob_batch = model_fnn_2(image_batch_scramble)
for i in range(10):
img = image_batch[i]
img_perm = image_batch_scramble[i]
real_label = label_batch[i].item()
log_pred_prob = log_pred_prob_batch[i]
# Output of the network are log-probabilities, need to take exponential for probabilities
pred_prob = torch.exp(log_pred_prob).data.numpy().squeeze()
visualize_pred(img_perm, pred_prob, real_label)
