Zero-shot prompting with reviews

As well as answering questions, transforming text, and generating new text, DeepSeek models can also be used for classification tasks, such as categorization and sentiment analysis.

In this exercise, you'll explore using DeepSeek's chat models for sentiment classification using reviews from an online shoe store called Toe-Tally Comfortable. Ideally, you're looking for outputs in the form of:

1. Review text = <1-5 rating>
2. Review text = <1-5 rating>
3. ...

Diese Übung ist Teil des Kurses

Working with DeepSeek in Python

Kurs anzeigen

Anleitung zur Übung

Define a prompt to classify the sentiment of the statements provided using the numbers 1 to 5 (positive to negative)— don't provide any example just yet!
Create a request to send this prompt to deepseek-ai/DeepSeek-V3.

Interaktive Übung

Vervollständige den Beispielcode, um diese Übung erfolgreich abzuschließen.

client = OpenAI(api_key="", base_url="https://api.together.xyz/v1")

# Define a multi-line prompt to classify sentiment
prompt = """____. Return no explanations:
1. Unbelievably good!
2. Shoes fell apart on the second use.
3. The shoes look nice, but they aren't very comfortable.
4. Can't wait to show them off!"""

# Create a request to the chat model
response = client.chat.completions.create(
  model="deepseek-ai/DeepSeek-V3",
  messages=[{"role": "user", "content": ____}],
  max_tokens=150
)

print(response.choices[0].message.content)

Code bearbeiten und ausführen