Per-Class Model Performance Metrics for XGBoost and a Fine-Tuned Model

Eui Jun Lee published a dataset on June 4, 2026, containing per-class performance metrics for two machine learning models. The data includes precision, recall, and F1 scores for XGBoost and a fine-tuned model, evaluated on a common test subset of 14,832 instances. The dataset is available under a CC-BY-4.0 license on the figshare platform.

Use Cases

• Compare the performance of XGBoost against a fine-tuned model based on per-class metrics.
• Analyze model strengths and weaknesses across different classes using precision, recall, and F1 scores.
• Benchmark classification model results on a test subset of 14,832 instances.
• Inform model selection decisions for classification tasks based on detailed evaluation metrics.

Strengths

• Dataset is small (5.5 KB), making it quick to download and inspect.
• Provides a direct comparison of two models (XGBoost and a fine-tuned model) on the same test subset.
• Includes three standard classification metrics (precision, recall, F1) for per-class analysis.

Limitations

• Row count and column-level documentation are unknown, which may limit suitability assessment.
• Description metadata is limited; actual data structure and semantics must be inferred after download.
• The dataset's tiny size (5.5 KB) indicates a very limited scope, likely containing only summary statistics.

Provenance

Source

figshare

Collection Method

Likely derived from a machine learning model evaluation experiment.

Freshness

Last updated 2026-06-04 17:25:21; freshness should be verified.