SKIPP'D is a dataset for short-term solar forecasting, combining sky images with photovoltaic power generation data. It was created by researchers Nie, Y., Li, X., Scott, A., Sun, Y., Venugopal, V., & Brandt, A. and published in Solar Energy in 2023. The dataset is hosted on Hugging Face by the organization solarbench.
Use Cases
- Training short-term solar forecasting models based on sky imagery and power generation data.
- Analyzing correlations between sky conditions and photovoltaic output.
- Benchmarking machine learning algorithms for renewable energy prediction.
- Developing computer vision models for solar irradiance estimation from sky images.
- Studying the impact of weather patterns on solar power generation.
Strengths
- Dataset is associated with a peer-reviewed publication in Solar Energy (2023).
- The description explicitly states it combines two data modalities: sky images and photovoltaic power generation.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- The dataset's freshness should be verified, as the last update timestamp is 2026-04-11.
Provenance
- Source
- solarbench