Claim · model-written summary
Model size decreased 36.8% to 9.1 M.
Not reportedTaxonMaterialGeneCompoundOrgan / stageInteractorInteractor roleSecond interactorSecond interactor roleTreatmentComparatorSetting
Verbatim
e0“model size was reduced by 36.8% (only 9.1 M)”
Title
YOLOv5s-T: A Lightweight Small Object Detection Method for Wheat Spikelet Counting
Lei Shi; Jiayue Sun; Yuanbo Dang; et al. Agriculture 2023. doi:10.3390/agriculture13040872
Other claims from this paper (10)
- Method / resource Predicted and true flowering stage spikelet counts had R2 of 0.97.
- Method / resource Predicted and true filling stage spikelet counts had R2 of 0.85.
- Method / resource Predicted and true mature stage spikelet counts had R2 of 0.78.
- Method / resource Across all three stages, R2 was 0.87.
- Method / resource Across all stages, RMSE was 0.70.
- Method / resource YOLOv5s-T spikelet detection and counting performance did not decline versus original YOLOv5s.
- Method / resource Training GPU memory use decreased by 0.82 GB.
- Method / resource Inference time decreased by 2.3 ms.
- Method / resource Processing time decreased by 10 ms.
- Method / resource Calculation amount also decreased.