Calibra: Open-source toolkit for robot dataset observability

Calibra is an open-source toolkit designed to help researchers audit robot datasets and identify quality issues before training policies. The first public release includes a Robot Dataset Health Check interactive Space for auditing LeRobot datasets and a benchmark of 30 public LeRobot datasets with health scores.

• Robot Dataset Health Check: An interactive Hugging Face Space for auditing LeRobot datasets in seconds.
• Benchmark of 30 public LeRobot datasets: Includes health scores, quality diagnostics, and community comparisons.
• Toolkit features: Quality-aware coreset selection, training outcome prediction, and a Python API/CLI.

The toolkit aims to address the difficulty of understanding dataset quality prior to training by providing auditing capabilities and outcome estimation.