Statistical Comparison of Batch Sizes Using 10-Fold Accuracy Values

A 5.5 KB Excel file containing statistical comparisons of different batch sizes against a best-performing model using 128 batch size. The dataset, authored by Ashik Mostafa Alvi, was last updated on April 20, 2026, and is shared under a CC-BY-4.0 license on figshare.

Use Cases

• Compare the performance of different batch sizes based on the 10-fold accuracy values mentioned in the description.
• Analyze the statistical significance of batch size selection for model training.
• Benchmark model accuracy against a reference model using a batch size of 128.
• Inform hyperparameter tuning strategies for neural network training.

Strengths

• Dataset is small (5.5 KB), facilitating quick download and inspection.
• Provides a direct statistical comparison against a defined best model (128 batch size).
• Uses a 10-fold cross-validation methodology, which suggests a robust evaluation approach.

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

figshare, author Ashik Mostafa Alvi.

Collection Method

Likely derived from machine learning model evaluation experiments.

Time Range

Creation or update timeframe not specified beyond the 2026 metadata date.

Freshness

Last updated 2026-04-20 17:58:12; freshness should be verified.

Geography

Spatial coverage is not specified.