Electricity Consumption Data: 2,000-Row Subsampled Benchmark

A 2,000-row, 100-column subsample of a larger electricity dataset, created by Eddie Bergman. It contains 10 target classes and was generated using a stratified sampling method with a fixed random seed. The original dataset is from OpenML and is in the public domain.

Use Cases

• Benchmarking classification algorithms based on the 10 target classes.
• Testing feature selection methods on the 100 available columns.
• Evaluating model performance on a controlled, stratified sample of electricity data.

Strengths

• Controlled sample size of 2,000 rows, which is suitable for rapid experimentation.
• Fixed random seed (3) ensures the sampling process is reproducible.
• Stratified sampling was used, which likely preserves the class distribution of the original data.

Limitations

• The specific features and their semantics are unknown, as column-level documentation is absent.
• Row count for the original source dataset is unknown, limiting assessment of the subsample's representativeness.
• Last update date is unknown; freshness unverified.

Provenance

Source

OpenML (dataset ID 44156), subsampled by Eddie Bergman.

Collection Method

Programmatic subsampling using a Python script with random column and row selection, and stratified sampling for the target variable.