Few-shot prompting: all the examples!
Now for the finale! Try adding the example Comfortable, but not very pretty = 2 to the prompt. As this example is more moderate, it might help the model with the troublesome second-to-last review.
This exercise is part of the course
Working with the OpenAI API
Exercise instructions
- Add the
Comfortable, but not very pretty = 2example to the prompt and re-run the request.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
client = OpenAI(api_key="")
# Add the final example
prompt = """Classify sentiment as 1-5 (negative to positive):
1. ____
2. Love these! = 5
3. Unbelievably good! =
4. Shoes fell apart on the second use. =
5. The shoes look nice, but they aren't very comfortable. =
6. Can't wait to show them off! = """
response = client.chat.completions.create(model="gpt-4o-mini", messages=[{"role": "user", "content": prompt}], max_completion_tokens=100)
print(response.choices[0].message.content)