G
News
March 16, 2023
4 min read

Giskard 1.4 is out! What's new in this version? ⭐

With Giskard’s new Slice feature, we introduce the possibility to identify business areas in which your AI models underperform. This will make it easier to debug performance biases or identify spurious correlations. We have also added an export/import feature to share your projects, as well as other minor improvements.

Giskard's turtle slicing some veggies!
Blanca Rivera Campos
Giskard's turtle slicing some veggies!
Giskard's turtle slicing some veggies!

We have released a new version 😻

Hi there,

This month we have the pleasure to announce that a new version 1.4 has been officially launched, hooray!

Check out the new version here.

🔪 Slice feature

This new feature will enable you to identify business areas in which your AI models underperform. Identifying these areas is critical because they can lead to catastrophic AI performance issues.

Create your own slice by simply adapting a Python function with the rows you might want to inspect. Once your slice is ready, you can easily ask for feedback to your business expert.

This feature will enable data scientists to debug performance biases or spurious correlations inside their model.

Slice feature - Giskard 1.4

📤 Export and import your projects

Now you can easily export and import your project (metadata, dataset, models, slices). This feature may come in handy for debugging and for sharing your project without having to share your instance.  

Export/import your projects - Giskard 1.4

🐞 Numerous bug fixes and minor improvements

For example:

  • Simplified user management by opening access without login by default.
  • Fixed ML Worker issues related to disconnecting after a while.

🗺 More to come

We’re already working in future releases 🔧

We have also started to work on how to test Generative language models...you will hear soon from us 👀

Thank you so much, and see you soon!