Abstract
Abstract Accurate and timely crop yield forecasts are crucial for guiding decisions on resource allocation, market strategies, and food security interventions, especially in food-insecure regions. These forecasts enable policymakers and stakeholders in the global food supply chain to anticipate production levels and implement effective mitigation and response measures. As the complexity of data and the need for precise predictions continue to grow, there is an increasing demand for models that can efficiently process and analyze diverse datasets. To address these challenges, we present an application of a foundation model for small- to medium-sized tabular data (TabPFN), to sub-national yield forecasting task in South Africa for maize, soybeans and sunflowers. We used the dekadal (10-day) time series of Earth Observation (EO; FAPAR and soil moisture) and gridded weather data (air temperature, precipitation and radiation) to forecast the yield of summer crops at the sub-national level. The crop yield data were available for 23 years and for up to 8 provinces. Covariate variables for TabPFN (i.e., EO and weather) were extracted by region and aggregated at a monthly scale. We benchmarked the results of the TabPFN against six ML models and two baseline models. Leave-one-year-out cross-validation experiment setting was used in order to ensure the assessment of the model’s capacity to forecast an unseen year. Results showed that TabPFN and ML models exhibit comparable accuracy, both outperforming the baseline approaches. For maize, the best-performing ML model achieved 6.8% rRMSEp, while TabPFN achieved 8.8%, with R² of 0.91 and 0.86 respectively at the national level. Nonetheless, TabPFN demonstrated practical utility due to its faster tuning time and reduced requirement for feature engineering.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 7
Claim · model-written summary
TabPFN and ML models showed comparable accuracy, both outperforming baseline approaches in sub-national yield forecasting.
- Taxon
- maize, soybeans and sunflowers
Glycine max
Helianthus
Zea mayse0 - Outcome
- accuracye2
- Direction
- outperforminge2
- Qualifier
- comparablee2
- Comparator
- baseline approachese2
- Setting
- South Africae0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“we present an application of a foundation model for small- to medium-sized tabular data (TabPFN), to sub-national yield forecasting task in South Africa for maize, soybeans and sunflowers”
e2“Results showed that TabPFN and ML models exhibit comparable accuracy, both outperforming the baseline approaches”
Claim · model-written summary
The best-performing ML model achieved 6.8% rRMSEp for maize at the national level.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e3“For maize, the best-performing ML model achieved 6.8% rRMSEp, while TabPFN achieved 8.8%, with R² of 0.91 and 0.86 respectively at the national level”
Claim · model-written summary
TabPFN achieved 8.8% rRMSEp for maize at the national level, higher than the best-performing ML model.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e3“For maize, the best-performing ML model achieved 6.8% rRMSEp, while TabPFN achieved 8.8%, with R² of 0.91 and 0.86 respectively at the national level”
Claim · model-written summary
The best-performing ML model reached R² of 0.91 for maize at the national level.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e3“For maize, the best-performing ML model achieved 6.8% rRMSEp, while TabPFN achieved 8.8%, with R² of 0.91 and 0.86 respectively at the national level”
Claim · model-written summary
TabPFN reached R² of 0.86 for maize at the national level.
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e3“For maize, the best-performing ML model achieved 6.8% rRMSEp, while TabPFN achieved 8.8%, with R² of 0.91 and 0.86 respectively at the national level”
Claim · model-written summary
TabPFN had faster tuning time than the six ML models it was benchmarked against.
- Taxon
- maize, soybeans and sunflowersGlycine maxHelianthusZea mayse0
- Outcome
- tuning timee4
- Direction
- fastere4
- Comparator
- six ML modelse1
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“we present an application of a foundation model for small- to medium-sized tabular data (TabPFN), to sub-national yield forecasting task in South Africa for maize, soybeans and sunflowers”
e1“We benchmarked the results of the TabPFN against six ML models and two baseline models”
e4“Nonetheless, TabPFN demonstrated practical utility due to its faster tuning time and reduced requirement for feature engineering”
Claim · model-written summary
TabPFN required reduced feature engineering compared with the six ML models it was benchmarked against.
- Taxon
- maize, soybeans and sunflowersGlycine maxHelianthusZea mayse0
- Outcome
- requirement for feature engineeringe4
- Direction
- reducede4
- Comparator
- six ML modelse1
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“we present an application of a foundation model for small- to medium-sized tabular data (TabPFN), to sub-national yield forecasting task in South Africa for maize, soybeans and sunflowers”
e1“We benchmarked the results of the TabPFN against six ML models and two baseline models”
e4“Nonetheless, TabPFN demonstrated practical utility due to its faster tuning time and reduced requirement for feature engineering”