MAE Loss for LSTM and MLP Models with Feature Selection Tests

A 5.5 KB Excel file contains MAE loss metrics for resource consumption of trained LSTM and MLP models. The dataset compares model performance with and without feature selection on a test set. It was authored by Iman Hindi and last updated in March 2026.

Use Cases

• Compare MAE loss between LSTM and MLP models for resource consumption prediction tasks.
• Analyze the impact of feature selection on model performance using the provided MAE loss metrics.
• Benchmark resource consumption model accuracy using the test set MAE loss data.

Strengths

• Dataset is small (5.5 KB), facilitating quick download and analysis.
• Provides a direct comparison of two model architectures (LSTM and MLP) under different conditions.
• Released under a permissive CC BY 4.0 license.

Limitations

• Extremely limited scope with only MAE loss metrics, lacking raw data, features, or row counts.
• Unknown sample size and column structure prevents assessment of statistical reliability.
• No contextual information about the resource types, test set composition, or training parameters.

Provenance

Source

figshare

Collection Method

Results from model training and evaluation experiments, methodology unspecified.

Time Range

null

Freshness

Last updated March 2026.

Geography

null