[모두를 위한 딥러닝 시즌2] Lab-07-2 MNIST Introduction

Posted Oct 17, 2024

By 띵석

views 7 min read

What is MNIST?

0~9까지 손으로 쓴 숫자로 구성된 데이터셋
우체국에서 우편번호를 자동으로 인식하기 위해 만들어짐
6만장 Train data, labes + 1만장 Test data, labels

Example of MNIST

28 x 28 이미지
1 channel gray image (흑백)
- 28 x 28 값으로 이루어짐
- 784개의 데이터
- pytorch 에서 X = X.view(-1, 28 * 28)를 이용해 사용
0 ~ 9 digits

torchvision

pytorch에서 사용하는 패키지
datasets : NMIST, IMAGENET 등 데이터 셋
models : Alexnet, VGG 등 모델 아키텍쳐
Transforms : 데이터에 적용할 수 있는 transform(전처리)
utils : 데이터를 쉽게 읽어오게 함

Reading data

  
import torchvision.datasets as dsets

...
mnist_train = dsets.MNIST(root="MNIST_data/", train=True, transform=transforms.ToTensor(), download=True)

mnist_test = dsets.MNIST(root="MNIST_data/", train=False, transform=transforms.ToTensor(), download=True)
...
# pytorch의 이미지는 0-1의 C,H,W 순서 값
# 일반적인 이미지는 0-255의 H,W,C 순서 값
# 이를 transforms.ToTenser()를 이용해 변환

data_loader = torch.utils.DataLoader(DataLoader=mnist_train, batch_size=batch_size, shuffle=True, drop_last=True)
# drop_last = batch size 만큼 데이터를 불러오고 남은 데이터를 가져올 지 설정

for epoch in range(training_epochs):
  ...
  for X, Y in data_loader:
  # X : MNIST 이미지
  # Y : label (0~9)
    X = X.view(-1, 28 * 28).to(device)
    # view를 이용해 (batch size, 1(channel), 28, 28) -> (batch size, 784) 로 변경
    
# epoch : Training set 전체를 학습하는 횟수
# batch size : data set을 나누는 크기
# iteration : batch를 이용해 학습하는 횟수

Softmax

  
# MNIST data image of shape 28 * 28 = 784
linear = torch.nn.Linear(784, 10, bias=True).to(device)

# parameters
training_epochs = 15
batch_size = 100

# define cost/loss & optimizer
criterion = torch.nn.CrossEntropyLoss().to(device)    # Softmax is internally computed.
optimizer = torch.optim.SGD(linear.parameters(), lr=0.1)

for epoch in range(training_epochs):
    avg_cost = 0
    total_batch = len(data_loader)

    for X, Y in data_loader:
        # reshape input image into [batch_size by 784]
        # label is not one-hot encoded
        X = X.view(-1, 28 * 28).to(device)
        Y = Y.to(device)

        optimizer.zero_grad()
        hypothesis = linear(X)
        cost = criterion(hypothesis, Y)
        cost.backward()
        optimizer.step()

        avg_cost += cost / total_batch

    print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))
    # Epoch: 0001 cost = 0.535468459
		# Epoch: 0002 cost = 0.359274179
		# Epoch: 0003 cost = 0.331187546
		# Epoch: 0004 cost = 0.316578031
		# Epoch: 0005 cost = 0.307158172
		# Epoch: 0006 cost = 0.300180733
		# Epoch: 0007 cost = 0.295130193
		# Epoch: 0008 cost = 0.290851533
		# Epoch: 0009 cost = 0.287417084
		# Epoch: 0010 cost = 0.284379542
		# Epoch: 0011 cost = 0.281825215
		# Epoch: 0012 cost = 0.279800713
		# Epoch: 0013 cost = 0.277808994
		# Epoch: 0014 cost = 0.276154280
		# Epoch: 0015 cost = 0.274440825

Test

  
# Test the model using test sets
with torch.no_grad(): # gradient 계산 x
    X_test = mnist_test.test_data.view(-1, 28 * 28).float().to(device)
    Y_test = mnist_test.test_labels.to(device)

    prediction = linear(X_test)
    correct_prediction = torch.argmax(prediction, 1) == Y_test
    accuracy = correct_prediction.float().mean()
    print('Accuracy:', accuracy.item())
    
		# Accuracy: 0.8862999677658081

Visualization

  
import matplotlib.pyplot as plt
import random

r = random.randint(0, len(mnist_test) - 1)
    X_single_data = mnist_test.test_data[r:r + 1].view(-1, 28 * 28).float().to(device)
    Y_single_data = mnist_test.test_labels[r:r + 1].to(device)

    print('Label: ', Y_single_data.item())
    single_prediction = linear(X_single_data)
    print('Prediction: ', torch.argmax(single_prediction, 1).item())

    plt.imshow(mnist_test.test_data[r:r + 1].view(28, 28), cmap='Greys', interpolation='nearest')
    plt.show()
    # Label:  8
		# Prediction:  3

실습 전체 코드

  
import torch
import torchvision.datasets as dsets
import torchvision.transforms as transforms
import matplotlib.pyplot as plt
import random
device = 'cuda' if torch.cuda.is_available() else 'cpu'

# for reproducibility
random.seed(777)
torch.manual_seed(777)
if device == 'cuda':
    torch.cuda.manual_seed_all(777)
# parameters
training_epochs = 15
batch_size = 100
# MNIST dataset
mnist_train = dsets.MNIST(root='MNIST_data/',
                          train=True,
                          transform=transforms.ToTensor(),
                          download=True)

mnist_test = dsets.MNIST(root='MNIST_data/',
                         train=False,
                         transform=transforms.ToTensor(),
                         download=True)
# dataset loader
data_loader = torch.utils.data.DataLoader(dataset=mnist_train,
                                          batch_size=batch_size,
                                          shuffle=True,
                                          drop_last=True)
# MNIST data image of shape 28 * 28 = 784
linear = torch.nn.Linear(784, 10, bias=True).to(device)
# define cost/loss & optimizer
criterion = torch.nn.CrossEntropyLoss().to(device)    # Softmax is internally computed.
optimizer = torch.optim.SGD(linear.parameters(), lr=0.1)
for epoch in range(training_epochs):
    avg_cost = 0
    total_batch = len(data_loader)

    for X, Y in data_loader:
        # reshape input image into [batch_size by 784]
        # label is not one-hot encoded
        X = X.view(-1, 28 * 28).to(device)
        Y = Y.to(device)

        optimizer.zero_grad()
        hypothesis = linear(X)
        cost = criterion(hypothesis, Y)
        cost.backward()
        optimizer.step()

        avg_cost += cost / total_batch

    print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost))

print('Learning finished')

# Test the model using test sets
with torch.no_grad():
    X_test = mnist_test.test_data.view(-1, 28 * 28).float().to(device)
    Y_test = mnist_test.test_labels.to(device)

    prediction = linear(X_test)
    correct_prediction = torch.argmax(prediction, 1) == Y_test
    accuracy = correct_prediction.float().mean()
    print('Accuracy:', accuracy.item())

    # Get one and predict
    r = random.randint(0, len(mnist_test) - 1)
    X_single_data = mnist_test.test_data[r:r + 1].view(-1, 28 * 28).float().to(device)
    Y_single_data = mnist_test.test_labels[r:r + 1].to(device)

    print('Label: ', Y_single_data.item())
    single_prediction = linear(X_single_data)
    print('Prediction: ', torch.argmax(single_prediction, 1).item())

    plt.imshow(mnist_test.test_data[r:r + 1].view(28, 28), cmap='Greys', interpolation='nearest')
    plt.show()

# Epoch: 0001 cost = 0.535150588
# Epoch: 0002 cost = 0.359577775
# Epoch: 0003 cost = 0.331264257
# Epoch: 0004 cost = 0.316404670
# Epoch: 0005 cost = 0.307106972
# Epoch: 0006 cost = 0.300456554
# Epoch: 0007 cost = 0.294933408
# Epoch: 0008 cost = 0.290956199
# Epoch: 0009 cost = 0.287074089
# Epoch: 0010 cost = 0.284515619
# Epoch: 0011 cost = 0.281914085
# Epoch: 0012 cost = 0.279526860
# Epoch: 0013 cost = 0.277636588
# Epoch: 0014 cost = 0.275874794
# Epoch: 0015 cost = 0.274422765
# Learning finished
# Accuracy: 0.8883000016212463
# Label:  8
# Prediction:  3

실습 코드 MNIST 오류상황

MNIST 다운로드 오류
- 홈페이지에 직접 들어가서 다운받아도 403 오류 발생
- 터미널에서 !wget [www.di.ens.fr/~lelarge/MNIST.tar.gz](http://www.di.ens.fr/~lelarge/MNIST.tar.gz) 와 !tar -zxvf MNIST.tar.gz 입력
- MNIST_data 폴더 안에 MNIST 데이터 직접 추가
MNIST_data/MNIST/processed/training.pt is a zip archive (did you mean to use torch.jit.load()?) 오류
- torch 1.6버전 이하에서 발생하는 문제
- pip install –upgrade torch 로 pytorch 업그레이드

Study, AI

This post is licensed under CC BY 4.0 by the author.

What is MNIST?

Example of MNIST

torchvision

Reading data

Softmax

Test

Visualization

Trending Tags