Setting up training arguments
After tokenizing your customer support interactions, you need to set up your training arguments before you can fine-tuned a pre-trained model.
TrainingArguments
has been loaded for you.
Este exercício faz parte do curso
Introduction to LLMs in Python
Instruções de exercício
- Set up an instance of
TrainingArguments()
. - Set the evaluation strategy as
"epoch"
. - Specify three training epochs.
- Set the batch sizes for both training and evaluation as three.
Exercício interativo prático
Experimente este exercício preenchendo este código de exemplo.
# Set up an instance of TrainingArguments
training_args = ____(
output_dir="./finetuned",
# Set the evaluation strategy
____,
# Specify the number of epochs
____=3,
learning_rate=2e-5,
# Set the batch sizes
____=3,
____=3,
weight_decay=0.01
)