Target Deployment Platforms: LiteFeatNet CNN Performance Metrics

Usman Rafi's dataset from 2026 contains performance metrics for the LiteFeatNet convolutional neural network. The model was trained and evaluated on 1,824 retinal fundus images from the Retinal Fundus Multi-Disease Image Dataset (RFMiD). It achieved a testing accuracy of 90.33% with an inference time of 4 milliseconds per image.

Use Cases

• Benchmarking lightweight CNN architectures based on reported performance metrics like accuracy and inference time.
• Evaluating model scalability for multi-class medical image classification based on the described framework.
• Assessing computational efficiency for clinical deployment based on model size (19.87 MB) and processing speed.

Strengths

• Performance metrics are explicitly reported, including 90.33% testing accuracy and 4 ms inference time.
• The model was validated on an external dataset (RFMiD 2.0) for generalizability.
• The dataset is small (5.5 KB), facilitating quick download and inspection.

Limitations

• The specific tabular data (columns and rows) are unknown; field semantics must be inferred after download.
• The dataset's scope is limited to performance metrics for a single model architecture.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Usman Rafi via figshare

Collection Method

Results from experiments training and evaluating the LiteFeatNet CNN.

Freshness

Last updated 2026-05-11 17:32:48; freshness should be verified.