Tokenizing text
You want to leverage a pre-trained model from Hugging Face and fine-tune it with data from your company support team to help classify interactions depending on the risk for churn. This will help the team prioritize what to address first, and how to address it, making them more proactive.
Prepare the training and test data for fine-tuning by tokenizing the text.
The data AutoTokenizer
and AutoModelForSequenceClassification
have been loaded for you.
Este exercício faz parte do curso
Introduction to LLMs in Python
Instruções de exercício
- Load the pre-trained model and tokenizer in preparation for fine-tuning.
- Tokenize both the
train_data["interaction"]
andtest_data["interaction"]
, enabling padding and sequence truncation.
Exercício interativo prático
Experimente este exercício preenchendo este código de exemplo.
# Load the model and tokenizer
model = ____.____("distilbert-base-uncased")
tokenizer = ____.____("distilbert-base-uncased")
# Tokenize the data
tokenized_training_data = ____(train_data["interaction"], return_tensors="pt", ____, ____, max_length=20)
tokenized_test_data = ____(test_data["interaction"], return_tensors="pt", ____, ____, max_length=20)
print(tokenized_training_data)