Eval Results: Robot Performance Benchmark from LeRobot

A dataset created using the LeRobot framework for evaluating robot performance. The dataset's structure includes metadata indicating a robot type of 'so_follower' and data organized into chunks. It was authored by zznzhang and last updated on June 20, 2026.

Use Cases

• Benchmarking robot control policies based on the evaluation framework mentioned in the description
• Analyzing robot task performance across different episodes and frames as suggested by the metadata structure
• Training or fine-tuning robot models using the collected evaluation trajectories

Strengths

• Data structure is defined with specific chunk sizes of 1000, facilitating organized access
• Metadata specifies a robot type ('so_follower'), providing context for the evaluation domain

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download
• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment

Provenance

Source

huggingface

Collection Method

Created using the LeRobot framework.

Freshness

Last updated 2026-06-20 11:41:31; freshness should be verified