how to setup epoch number in SFTTrainer?

[KlausikPL]

there my example code
from datasets import load_dataset
from trl import SFTTrainer

dataset = load_dataset("IMDB", split="train")

trainer = SFTTrainer(
"sshleifer/tiny-gpt2",
train_dataset=dataset,
dataset_text_field="text",
max_seq_length=512,
)
trainer.train()

[younesbelkada]

You need to instantiate a new object of TrainingArguments and pass num_train_epochs to it. Check out for example what is done here: https://github.com/huggingface/trl/blob/main/examples/scripts/sft.py#L113

[KlausikPL]

thanks you for help