Performance Comparison of 1D-CNN, AccuSleep, and DeepSleepNet for Sleep Stage Validation

9.5 KB of tabular data comparing the performance of three deep learning models—1D-CNN, AccuSleep, and DeepSleepNet—for sleep stage classification. The dataset, authored by Jinyoung Choi and last updated on April 23, 2026, includes results from within-subject and cross-subject validation using raw, z-scored, and mixture z-scored data. It is hosted on figshare under a CC-BY-4.0 license.

Use Cases

• Benchmarking 1D-CNN, AccuSleep, and DeepSleepNet models based on the performance metrics provided.
• Comparing within-subject versus cross-subject validation strategies for sleep analysis.
• Evaluating the impact of data normalization (raw, z-scored, mixture z-scored) on model performance.
• Informing methodological choices for sleep stage classification studies.

Strengths

• Provides a direct comparison of three established deep learning models for a specific task.
• Includes results for multiple validation strategies and data preprocessing methods.
• Dataset is small (9.5 KB) and likely easy to inspect and analyze.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset's scope is limited to performance metrics, not the underlying sleep data.

Provenance

Source

figshare

Collection Method

Likely contains aggregated performance metrics from computational experiments.

Time Range

The temporal coverage of the underlying experiments is not specified.

Freshness

Last updated 2026-04-23 17:25:29; freshness should be verified.

Geography

Spatial coverage is not specified.