ft.py

from simpletransformers.language_modeling import LanguageModelingModel
import logging


logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)

train_args = {
    "reprocess_input_data": True,
    "overwrite_output_dir": True,
    "train_batch_size": 64,
    "num_train_epochs": 3,
    "mlm": False,
}

model = LanguageModelingModel('gpt2', 'gpt2', args=train_args)

model.train_model("data/train.txt", eval_file="data/test.txt")

model.eval_model("data/test.txt")