Auto labeling using GPT-4

[imenelydiaker]

• Prompt for labeling with GPT-4 variants
• Script to compare labels from GPT with human annotations
• Update all entries with username: "auto" or "gpt"

Summary by Sourcery

This pull request adds a new script for pre-labeling datasets using GPT-4. The script downloads metadata and votes, processes images, generates labels using the OpenAI API, and saves or uploads the results to the Hugging Face Hub.

• New Features:
  • Introduced a script for pre-labeling the dataset using GPT-4, which includes functionality to download metadata and votes, process images, and generate labels using the OpenAI API.
• Enhancements:
  • Added functionality to save and upload metadata and votes to the Hugging Face Hub.

[sourcery-ai]

Reviewer's Guide by Sourcery

This pull request introduces a new script auto_labeling_using_llm.py to automate the pre-labeling of datasets using GPT-4. The script includes functions for handling metadata, retrieving votes, computing concepts, and making requests to the OpenAI API. The main function orchestrates the entire labeling process and saves the results locally or pushes them to the hub.

File-Level Changes

Files	Changes
scripts/auto_labeling_using_llm.py	Introduced a new script to automate dataset labeling using GPT-4, including functions for metadata handling, OpenAI API requests, and result processing.

---

Tips

• Trigger a new Sourcery review by commenting @sourcery-ai review on the pull request.
• You can change your review settings at any time by accessing your dashboard:
  • Enable or disable the Sourcery-generated pull request summary or reviewer's guide;
  • Change the review language;
• You can always contact us if you have any questions or feedback.

[sourcery-ai]

Hey @imenelydiaker - I've reviewed your changes and they look great!

Here's what I looked at during the review

• 🟡 General issues: 5 issues found
• 🟢 Security: all looks good
• 🟢 Testing: all looks good
• 🟢 Complexity: all looks good
• 🟢 Documentation: all looks good

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment to tell me if it was helpful.}

[sourcery-ai]

issue: Unused import 'random'

The 'random' module is imported but not used anywhere in the script. Consider removing it to keep the code clean.

[sourcery-ai]

issue (bug_risk): Potential off-by-one error

The slicing operation might exclude the closing brace '}'. Consider using 'pred.rfind("}") + 1' to include it.

[Xmaster6y]

Additionally add a image limit to debug, like only labelling the 10 first (or random) images