Fine-tuning for review classification
Your project continues and you now want to classify the reviews as positive, neutral, or negative. You'll use a fine-tuned model to automatically categorize the reviews.
The TrainingArguments and Trainer classes have been imported from transformers.
The training_args, model, and tokenized_datasets have also been pre-imported.
This exercise is part of the course
Reinforcement Learning from Human Feedback (RLHF)
Exercise instructions
- Create the train and test datasets using
"train"and"test"subsets from thetokenized_datasetsdataset. - Initialize the trainer class.
- Add training arguments to the class.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Define the train and test datasets
training_dataset = ____
testing_dataset = ____
# Initialize the trainer class
trainer = ____(
# Add arguments to the class
model=model,
args=training_args,
train_dataset=____,
eval_dataset=____
)