Absolute-Value

import torch
import torch.nn as nn
import torch.optim as optim
from sklearn.datasets import load_iris, load_breast_cancer, load_digits, load_wine
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler

dataset_name = "wine" # iris, breast_cancer, digits, wine
hidden_dim = 10
epochs = 50

Absolute-Value

import torch
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

# GPUが使える場合はGPUを使用
　if torch.cuda.is_available():
  device = 'cuda'
else:
  device = 'cpu'
print(f'device : {device}')