Install required libraries

!pip install transformers datasets

import torch from transformers import AutoTokenizer, AutoModelForSequenceClassification, Trainer, TrainingArguments from datasets import load_dataset

Cek apakah GPU tersedia

device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")

Load dataset

dataset = load_dataset("imdb")

Load pre-trained tokenizer dan model

model_name = "stepfun-ai/GOT-OCR2_0" tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForSequenceClassification.from_pretrained(model_name, num_labels=2) model.to(device)

Tokenize dataset

def preprocess_function(examples): return tokenizer(examples['text'], truncation=True, padding=True)

tokenized_datasets = dataset.map(preprocess_function, batched=True)

Split dataset into train and test

train_dataset = tokenized_datasets["train"].shuffle(seed=42).select(range(10000)) # menggunakan subset kecil untuk contoh test_dataset = tokenized_datasets["test"].shuffle(seed=42).select(range(2000))

Definisikan argumen pelatihan

training_args = TrainingArguments( output_dir="./results", evaluation_strategy="epoch", per_device_train_batch_size=8, per_device_eval_batch_size=8, num_train_epochs=3, weight_decay=0.01, )

Definisikan Trainer

trainer = Trainer( model=model, args=training_args, train_dataset=train_dataset, eval_dataset=test_dataset, )

Melatih model

trainer.train()

Evaluasi model

trainer.evaluate()

Simpan model setelah dilatih

model.save_pretrained("./trained_model") tokenizer.save_pretrained("./trained_model")