Data Augmentation데이터 증강

import torchvision.transforms as T
import torchvision.transforms.v2 as T2

# 기본 증강
basic_transform = T.Compose([
    T.RandomHorizontalFlip(p=0.5),
    T.RandomVerticalFlip(p=0.1),
    T.RandomRotation(degrees=15),
    T.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.2, hue=0.1),
    T.RandomResizedCrop(224, scale=(0.8, 1.0)),
    T.ToTensor(),
    T.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

# 고급 증강 (AutoAugment)
auto_transform = T.Compose([
    T.AutoAugment(T.AutoAugmentPolicy.IMAGENET),
    T.ToTensor()
])

# RandAugment: 랜덤하게 N개 변환 선택 후 magnitude M으로 적용
rand_transform = T.Compose([
    T.RandAugment(num_ops=2, magnitude=9),
    T.ToTensor()
])

import torch
import numpy as np

def mixup(x, y, alpha=0.2):
    """두 샘플을 선형 보간"""
    lam = np.random.beta(alpha, alpha)
    idx = torch.randperm(x.size(0))
    x_mix = lam * x + (1 - lam) * x[idx]
    y_a, y_b = y, y[idx]
    return x_mix, y_a, y_b, lam

def mixup_criterion(criterion, pred, y_a, y_b, lam):
    return lam * criterion(pred, y_a) + (1 - lam) * criterion(pred, y_b)

def cutmix(x, y, alpha=1.0):
    """이미지 일부를 다른 이미지로 교체"""
    lam = np.random.beta(alpha, alpha)
    idx = torch.randperm(x.size(0))
    _, _, H, W = x.shape
    cut_ratio = np.sqrt(1 - lam)
    cut_h, cut_w = int(H * cut_ratio), int(W * cut_ratio)
    cx, cy = np.random.randint(W), np.random.randint(H)
    x1 = max(cx - cut_w // 2, 0); x2 = min(cx + cut_w // 2, W)
    y1 = max(cy - cut_h // 2, 0); y2 = min(cy + cut_h // 2, H)
    x_mix = x.clone()
    x_mix[:, :, y1:y2, x1:x2] = x[idx, :, y1:y2, x1:x2]
    lam_real = 1 - (x2-x1)*(y2-y1)/(W*H)
    return x_mix, y, y[idx], lam_real

import random

def text_augment(text, method='swap'):
    words = text.split()
    if method == 'swap' and len(words) > 1:
        i, j = random.sample(range(len(words)), 2)
        words[i], words[j] = words[j], words[i]
    elif method == 'delete':
        idx = random.randint(0, len(words) - 1)
        words.pop(idx)
    elif method == 'duplicate':
        idx = random.randint(0, len(words) - 1)
        words.insert(idx, words[idx])
    return ' '.join(words)

# Back-translation: 한→영→한으로 번역 후 원문 대체