Skip to content
TheChatArchiveThe Chat Archive
o3-mini· OpenAIHugging Face LMArena Human Preference 140K

I'm using vscode, python, and jupyter for interactive python. I don't like the built in…

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with Amazon Novaamazon.nova-pro-v1:0’s answer.
  1. User

    I'm using vscode, python, and jupyter for interactive python. I don't like the built in data viewer very much and Data Wrangler is also kind of meh. In particular these don't seem to all 'drilling down' into arbitrary python objects or data structures. I'm looking for something that functions more along the lines of the built in data viewer in PyCharm. Any way to accomplish this?

  2. Assistant

    There isn’t a “silver‐bullet” VS Code extension that exactly mimics PyCharm’s data viewer with deep “drill‐down” capability into arbitrary objects, but there are a few approaches and workarounds you might try:

    1. VS Code’s Built-In Variable/Data Viewers:
      VS Code’s interactive window and debugger come with a variable explorer and a built‑in data viewer that works pretty well for many DataFrames and arrays. However, these tools tend to be tuned toward specific types (like pandas DataFrames) and don’t always offer the deep object exploration that PyCharm provides.

    2. Third-Party Libraries:
      If your interest is mostly in exploring pandas-like objects, you might consider tools that launch a separate, interactive viewer. Two popular ones are:

      • PandasGUI: A graphical user interface for exploring DataFrames that lets you filter, sort, and even chart your data.
      • DTale: Which creates a web-based interface from your Python session to inspect and analyze data interactively.

      Both open a separate window (or browser tab) and can provide a more “PyCharm-like” data exploration experience with some extra interactivity.

    3. Jupyter Notebook Widgets:
      For in-notebook exploration, you might check out widgets such as qgrid. Although qgrid is generally focused on DataFrames, it gives you a hands-on interactive grid that lets you sort, filter, and drill into the details. (Keep in mind that support and features are a bit different from PyCharm’s viewer and may not cover completely arbitrary objects.)

    4. Feature Requests and Future Outlook:
      There are ongoing discussions on the VS Code Python extension GitHub issues about enhancing the variable explorer (and data viewer) to better mimic PyCharm’s experience. It might be worth keeping an eye on those threads or even contributing your feedback there.

    5. Combining Tools:
      Some users find that by combining VS Code’s built‑in debuggers (using the watch panel for nested objects) with one of the external GUI tools above, they can cover most of their interactive exploration needs.

    Conclusion:
    Right now, if your workflow calls for deep diving into arbitrary Python objects the way PyCharm’s data viewer does, using a specialized GUI tool (like PandasGUI or DTale) alongside VS Code might be your best bet. Alternatively, keep an eye on future updates in the VS Code ecosystem—the community is actively working on enhancing interactive features, and we might see tools that close the gap with PyCharm in coming releases.

    I hope this helps you find a workflow that suits your needs!

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.