YieldLearn: An interpretable machine-learning pipeline for estimating yield-related traits based on physiological indicators
- 1National Research Council Canada
Abstract
Efficient yield prediction under field conditions remains a key challenge for crop breeders and agronomists. YieldLearn is an interpretable machine-learning pipeline developed to estimate yield-related traits in durum wheat (Triticum durum) using physiological indicators measured by the Hansatech Instruments Pocket PEA. The workflow integrates fluorescence parameters (Area, Fo, Fm, Fv/Fm, PI Inst.) with field-measured agronomic traits (Booting, Height, Maturity, Lodging, Yield kg/plot, and Adjusted Yield kg/ha). Five supervised learning algorithms multinomial logistic regression, random forest, support vector machine (radial kernel), k-nearest neighbors, and decision tree were evaluated using five-fold cross-validation. Models achieved average classification accuracy around 0.57 with AUC values exceeding 0.58. SHAP (SHapley Additive exPlanations) visualizations revealed that high Fv/Fm and PI Inst. values were positively associated with yield potential, whereas elevated Fo indicated stress-induced yield reduction. An interactive Shiny App enables real-time prediction and feature-importance exploration. Given the small sample size (165 plots) and absence of environmental covariates such as temperature or soil moisture, YieldLearn should be regarded as a reproducible exploratory framework rather than a production-ready predictor. Nonetheless, it offers a valuable foundation for guiding experimental design, validating physiological indicators, and supporting the next generation of scalable, interpretable crop-yield models.
Related
- An interpretable multimodal deep learning model for physiological stress detection
- Comparative field study on water deficit effects on morphological and physiological indicators in durum wheat
- Physiological Adaptations and Thermoregulatory Indicators of Heat Stress in Bovine Production Systems.
- Mitigation of drought stress in lettuce (Lactuca sativa L.) through application of zinc-oxide nanoparticles and humic acid: evaluation of growth, physiological, and biochemical indicators
- Machine learning and spatial regression approaches to estimating willingness to pay for ecosystem services in Jakarta and Taoyuan
BibTeX
@misc{zhang2026yieldlearn,
title = {{YieldLearn: An interpretable machine-learning pipeline for estimating yield-related traits based on physiological indicators}},
author = {X. Zhang and H. Shaterian and P. Ashe},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.22.753232},
url = {https://www.biorxiv.org/content/10.64898/2026.09.22.753232v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - YieldLearn: An interpretable machine-learning pipeline for estimating yield-related traits based on physiological indicators
AU - X. Zhang
AU - H. Shaterian
AU - P. Ashe
PY - 2026
DA - 2026/09/22
PB - bioRxiv
DO - 10.64898/2026.09.22.753232
UR - https://www.biorxiv.org/content/10.64898/2026.09.22.753232v1
AB - Efficient yield prediction under field conditions remains a key challenge for crop breeders and agronomists. YieldLearn is an interpretable machine-learning pipeline developed to estimate yield-related traits in durum wheat (Triticum durum) using physiological indicators measured by the Hansatech Instruments Pocket PEA. The workflow integrates fluorescence parameters (Area, Fo, Fm, Fv/Fm, PI Inst.) with field-measured agronomic traits (Booting, Height, Maturity, Lodging, Yield kg/plot, and Adjusted Yield kg/ha). Five supervised learning algorithms multinomial logistic regression, random forest, support vector machine (radial kernel), k-nearest neighbors, and decision tree were evaluated using five-fold cross-validation. Models achieved average classification accuracy around 0.57 with AUC values exceeding 0.58. SHAP (SHapley Additive exPlanations) visualizations revealed that high Fv/Fm and PI Inst. values were positively associated with yield potential, whereas elevated Fo indicated stress-induced yield reduction. An interactive Shiny App enables real-time prediction and feature-importance exploration. Given the small sample size (165 plots) and absence of environmental covariates such as temperature or soil moisture, YieldLearn should be regarded as a reproducible exploratory framework rather than a production-ready predictor. Nonetheless, it offers a valuable foundation for guiding experimental design, validating physiological indicators, and supporting the next generation of scalable, interpretable crop-yield models.
ER -