OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Data & Analytics

Spotlight

Spotlight is an open-source Python developer tool that turns dataframe-based unstructured datasets into interactive browser views for exploring examples, patterns and model-related data.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Spotlight — OpenViST project profile in AI Data & Analytics.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Spotlight when your data is available through a dataframe or supported dataset and you need an interactive view for examining examples, enrichments or model-debugging information.

Who it is for

  • Machine-learning teams examining data quality and model outputs
  • Engineers working with unstructured datasets
  • Developers who want to inspect image, audio, text, video, time-series or geometric data alongside dataframe fields

Documented uses

  • Visualize unstructured records held in a dataframe.
  • Use embeddings, predictions and uncertainty information to inspect patterns and identify clusters.
  • Review classification examples with labels, predictions, embeddings and selected features in a debugging layout.
  • Explore and communicate data issues across supported modalities.
Before you choose

Requirements & limits

The project is open source under the MIT license and documents installation through pip.

  • The documented quickstart requires Python 3.10 or later.
  • Optional analyzers and embeddings require the analyzers and torch package extras, respectively.
  • Hugging Face datasets version 4 audio workflows require the torch extra and FFmpeg.
  • The README documents crash-report and performance collection and provides SPOTLIGHT_OPT_OUT=true as an opt-out for crash reporting.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Open a dataframe in Spotlight

Install Spotlight and open a dataframe for interactive exploration in a browser.

Before you start

  • Python 3.10 or later
  • A pandas dataframe, or a CSV that can be loaded into one
  • The renumics-spotlight package installed with pip

Steps

  1. Install the package with the command below.

    Documented command
    pip install renumics-spotlight

  2. Load a CSV into a pandas dataframe.

  3. Import spotlight from renumics and call spotlight.show(df, dtype={"image": spotlight.Image}) when the dataframe includes an image column.

Expected outcome

Spotlight opens in a browser with the dataframe ready to explore; the dtype argument identifies the image column for the viewer.

Guide limitations

  • The documented example uses a CSV and pandas dataframe; other data-loading workflows may require their own dependencies.
  • A separate virtual environment is recommended by the project.
Documentation 1 ↗
Source-based practical guide

Explore a Hugging Face audio dataset with a debugging layout

Load the documented enriched audio dataset and display it with a predefined model-debugging layout.

Before you start

  • Python 3.10 or later
  • Spotlight with the torch extra
  • The Hugging Face datasets[audio] package
  • FFmpeg when using Hugging Face datasets version 4

Steps

  1. Install Spotlight with the documented torch extra.

  2. Load renumics/emodb-enriched with the Hugging Face datasets package using the all split.

  3. Create a debugging layout with spotlight.layouts.debug_classification, setting the label, prediction, embedding and feature columns shown in the documented example.

  4. Call spotlight.show(ds, layout=layout).

Expected outcome

Spotlight opens the loaded dataset with a predefined layout for inspecting classification data.

Guide limitations

  • The example depends on the named dataset and its documented columns.
  • For Hugging Face datasets version 4, the README specifies FFmpeg and the torch extra for audio use.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
Renumics/spotlight · Official homepage title · retained in its source language
How we review sourcesSuggest a correction